Turtles and pterosaurs tested again, nine years later with 1300 more taxa

In 2011,
back when the large reptile tree (LRT, 1786+ taxa) was much smaller (only about 400 taxa) I attempted a rather odd test. I wondered if turtles and pterosaurs (both Lepidosauromorpha in the LRT) would nest together while playing taxon deletion games. Here’s the 2011 link:

https://pterosaurheresies.wordpress.com/2011/07/25/pterosaurs-and-turtles-say-it-aint-so/

Click here to see those 2011 turtle-pterosaur results, still posted online.

Other workers interested in pterosaurs
(most recently Ezcurra et al. 2020) also like to play taxon deletion games as they attempt to cherry-pick preferred sisters close to dinosaurs while omitting tested and validate sisters far from dinosaurs.

The backstory
Peters 2000 added four tritosaur tanystropheid, fenestrasaur pterosaur precursors, Langobardisaurus, Cosesaurus, Sharovipteryx, and Longisquama (Fig. 1) to four previously published analyses and in every case these four nested closer to pterosaurs than any archosaur, archosauriform or archosauromorph. Unfortunately those taxa were omitted from more recent analysis, like those of Kellner 2003, Unwin 2003, Hone and Benton 2007, 2008, Bennett 2012 and Ezcurra et al. 2020.

A few years later, but still 14 years ago,
Peters 2007 added the lepidosaur, Huehuecuertzpalli and it attracted the four fenestrasaurs + pterosaurs. The LRT nested turtles within the Lepidosauromorpha here in 2011, updated here in 2014.

Now that many more taxa are present in the LRT,
let’s rerun that test and its various deletion subunits.

Today, in 2020, repeating the experiment with more taxa
deleting all lepidosauromorphs, other than turtles (and their ancestors back to Stephanospondylus) and pterosaurs, and keeping all archosauromorphs and enaliosaurs. Outgroups retained = Gephyrostegus and Silvanerpeton.

Results: Pterosaurs nest with turtles and basal sea reptiles rather than archosaurs and archosauromorphs.

Adding back all basal diapsids and protosaurs

Results: Basal diapsids as the first large clade, followed by protorosaurs with tritosaurs based on the convergence found there.

Adding back all tritosaurs
(= Macrocnemus as the last common ancestor) nests turtles and pareiasaurs as the first large clade, tritosaurs (including pterosaurs) as the next large clade, followed by archosauromorophs (including Lagerpeton and Scleromochlus).

Figure 1. Squamates, tritosaurs and fenestrasaurs in the phylogenetic lineage preceding the origin of the Pterosauria.

Deleting all non-fenestrasaur tritosaurs
(= Cosesaurus as the last common ancestor)

Results: Fenestrasaurs nest with basal diapsids. Orovenator is the proximal outgroup.

Deleting all non-pterosaur fenestrasaurs
(Bergamodactylus as the last common ancestor).

Results: Pterosaurs nest between turtles and choristoderes.

Figure 2. Basal diasids and proto-diapsids. Largely ignored these putative synapsids actually split from other synapsids while retaining the temporal fenestra trait that serves as the basis for the addition of upper temporal fenestra in diapsids. Included here are Protorothyris, Archaeovenator, Mycterosaurus, Heleosaurus, Mesenosaurus, Broomia, Milleropsis, Eudibamus, Petrolacosaurus, Spinoaequalis, and Tangasaurus.

Deleting all basal diapsids
(only turtles, pterosaurs and archosauromorphs are ingroup taxa).

Results: Pterosaurs nest between turtles and choristoderes, far from Scleromochlus, dinosaurs and Lagerpeton.

Deleting all turtle ancestors 
(= deleting Stephanospondylus through pareiasaurs)

Results: Pterosaurs nest between turtles and choristoderes, far from Scleromochlus, dinosaurs and Lagerpeton.

Figure 3. Faxinalipterus matched to Scleromochlus. The former is more primitive, like Gracilisuchus, in having shorter hind limbs and more robust fore limbs. The maxilla with fenestra and fossa, plus the teeth, are a good match.

Re-inserting terrestrial younginiforms and protorosaurs
to this last taxon list.

Results: Two large clades follow the turtle clade. Pterosaurs nest between three basal Youngina specimens and the clade Protorosauria, apart from the terrestrial younginiformes (other Youngina specimens + Pararchosauriformes (= Proterosuchus as the last common ancestor and choristoderes) and Euarchosauriformes (= Euparkeria as the last common ancestor and Lagerpeton and Scleromochlus).

Figure 4. The closest kin of Tropidosuchus are the much larger Chanaresuchus (matching Nesbitt 2011) and the smaller Lagerpeton.

Bottom line:
With or without tritosaurs and fenestrasaurs, pterosaurs prefer to nest with terrestrial younginiforms, choristoderes or turtles rather than lagerpetids, dinosaurs or Scleromochlus. Taxon exclusion remains the problem in traditional cladograms (like the recent Ezcurra et al. 2020).

Please send this post to anyone who still believes or protects
the outmoded clades ‘Ornithodira’ or ‘Avemetatarsalia’. Too many professors and their students are clinging to invalidated myths based on taxon exclusion — which is not what real scientists do. Real scientists test all competing candidates without cherry-picking or omitting taxa to suit their personal whims and traditions, in fear of their professors or colleagues.

If you would like to play taxon deletion games with the LRT,
click here, then click on the yellow CLICK HERE for LRT MacClade.nex file box with your request.

---

References
Bennett SC 1996. The phylogenetic position of the Pterosauria within the Archosauromorpha. Zoological Journal of the Linnean Society 118:261-308.
Bennett SC 2012. The phylogenetic position of the Pterosauria within the Archosauromorpha re-examined. Historical Biology. iFirst article, 2012, 1–19.
Benton MJ 1983. The Triassic reptile Hyperodapedon from Elgin, functional morphology and relationships. Philosophical Transactions of the Royal Society of London, Series B, 302, 605-717.
Brusatte SL , Benton MJ , Desojo JB and Langer MC 2010. The higher-level phylogeny of Archosauria (Tetrapoda: Diapsida), Journal of Systematic Palaeontology, 8:1, 3-47.
Evans SE 1988. The early history and relationships of the Diapsida. In: M. J. BENTON (Ed.), The Phylogeny and Classificationof the Tetrapoda. 1. Amphibians, Reptiles, Birds. Systematics Symposium Association Special Volume; Oxford (Clarendon Press), 221–260.
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
Hone DWE and Benton MJ 2007. An evaluation of the phylogenetic relationships of the pterosaurs to the archosauromorph reptiles. Journal of Systematic Palaeontology 5:465–469.
Hone DWE and Benton MJ 2008. Contrasting supertree and total evidence methods: the origin of the pterosaurs. Zitteliana B28:35–60.
Jalil N-E 1997. A new prolacertiform diapsid from the Triassic of North Africa and the interrelationships of the Prolacertiformes. Journal of Vertebrate Paleontology 17(3), 506-525.
Kellner AWA 2003. Pterosaur phylogeny and comments on the evolutionary history of the group. Geological Society Special Publications 217: 105-137.
Nesbitt SJ 2011. The early evolution of archosaurs: relationships and the origin of major clades. Bulletin of the American Museum of Natural History 352: 292 pp.
Peters D 2000b. A Redescription of Four Prolacertiform Genera and Implications for Pterosaur Phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106 (3): 293–336.
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.
Sereno PC 1991. Basal archosaurs: phylogenetic relationships and functional implications. Journal of Vertebrate Paleontology 11 (Supplement) Memoire 2: 1–53.
Unwin DM 2003. On the phylogeny and evolutionary history of pterosaurs. Pp. 139-190. in Buffetaut, E. & Mazin, J.-M., (eds.) (2003). Evolution and Palaeobiology of Pterosaurs. Geological Society of London, Special Publications 217, London, 1-347.

https://pterosaurheresies.wordpress.com/2012/09/27/bennett-2012-still-barking-up-the-wrong-pterosaur-tree/

https://pterosaurheresies.wordpress.com/2020/12/10/new-pterosaur-precursor-study-excludes-all-pterosaur-precursors/

https://pterosaurheresies.wordpress.com/2012/04/13/a-supertree-of-pterosaur-origins-hone-and-benton-2007-2009/

Another turtle origin paper omits critical taxa, inserts irrelevant taxa

de la Fuente et al. 2020 focus on
Triassic hard-shell turtles from South America.

Highlights from de la Fuente et al.
“We report new information about the oldest turtles from South America.”

Unfortunately, the oldest turtles are not the most primitive. Joyce 2007 wrote: “Currently there is little reason to doubt that Proganochelys quenstedti Bauer from the Upper Triassic of Europe is the most primitive well-understood representative of Testudinata.”

That has been wrong since 2016. Adding taxa to the large reptile tree (LRT, 1740+ taxa) sheds new light on the dual origin of turtles, with soft-shell turtles distinct from and parallel to hard-shell turtles from different small, horned pareiasaurs. We didn’t know that in 2007, but we’ve known this for the last four years, so de al Fuente is working with old data and traditions.

Highlights continued:
“We discuss the putative ichnological Triassic record of turtles.”

Turtle digits making imprints are interesting, especially since turtle arms extend anteriorly, not laterally, as in more primitive hard-shell turtles like Meiolania (Fig. 1) and most tetrapods.

Figure 1. Meiolania, the most primitive of known turtles, has lateral forelimbs, like non turtles.

Highlights continued:
“Triassic testudinatans show different strategies in building up the shell periphery.”

But if they don’t include the most primitive hard-shell turtles, Niolamia (Fig. 3) and Meiolania (Fig. 1), then de al Fuente et al. 2020 are not starting from the beginning, nor do they employ valid outgroups.

Highlights continued:
“We discuss new perspectives on the origin and early evolution of the turtle shell.”

Those new perspectives include irrelevant turtle-mimic taxa, like Pappochelys and Eorhynchochelys and omit all pareiasaurs and pre-pareiasaurs, Stephanospondylus and Carbonodraco. Without a valid phylogenetic contact and foundation, the whole study is somewhere between shaky and doomed.

Figure 2. Odontochelys is a basal soft-shell turtle with teeth and anterior elbows and extremely pronated forelimbs.

de la Fuente et al. consider
Odontochelys semitestacea (Fig. 2) from the Carnian of China, “the undisputed sister taxon” of turtles.

This is only half right. Adding taxa shows Odontochelys is basal to soft-shell turtles only. Sclerosaurus and Arganaceras are outgroup taxa not mentioned by de la Fuente et al. Hard-shell turtles have a different lineage arising from pareiasaurs (Fig. 3) which was omitted or overlooked by de al. Fuente et al. 2020.

de la Fuente et al. continue
“and some putative successive sister group taxa were found in Middle and Late Triassic rocks (e.g., Pappochelys rosinae from the Ladinian of Germany, Eorhynchochelys yuantouzhuensis from the Carnian of China).”

Adding taxa shows Pappochelys is a basal placodont. Eorhynchchelys is closer to Acleistorhinus. I’m still wondering when turtle workers are going to add taxa to confirm what the LRT recovered by minimizing taxon exclusion, rather than purposefully cherry-picking traditional taxa and omitting taxa tested against all other candidates and therefore shown to be valid.

Figure 3. Bunostegos and Elginia at the base of hard shell turtles in the LRT, where late-surviving Niolamia and Meiolania are basal hardshell turtles more primitive than Proganochelys and Palaeochesis. Not all colors here match those in figure 1. Note the labels.

de la Fuente et al. continue
“These three facts highlight the importance of the Triassic for the origin and early evolution of turtles.”

These are not facts. Simply adding taxa shows these were untested hypotheses falsified in 2016 by taxon inclusion. Let the software do what it does best. Keep your biases, traditions and cherry-picking out of it.

Figure x. Carbonodraco enters the LRT alongside another recent addition, Kudnu, at the base of the pareiasaurs + turtles.

Latest additions to the LRT
Recently we looked at Palaeochersis, a third Triassic turtle derived from the late-surviving, most primitive turtles with horns (Fig. 3), not the traditional other way around. Horned turtles are also from South America, but were overlooked by de la Fuente, Sterli and Krapovickas 2020 because they believed traditional myths invalidated in 2016,

---

References
de la Fuente MS, Sterli J and  Krapovickas V 2020. Triassic turtles from Pangea: The legacy from South America. Journal of South American Earth Sciences 102910
doi: https://doi.org/10.1016/j.jsames.2020.102910
https://www.sciencedirect.com/science/article/abs/pii/S0895981120304533
Joyce WG 2007. Phylogenetic relationships of Mesozoic turtles. Bulletin of the Peabody Museum of Natural History 48(1):3–102..

dual origin of turtles paper on ResearchGate.net

Abstract from The Dual Origin of Turtles from Pareiasaurs
“The origin of turtles (traditional clade: Testudines) has been a vexing problem in paleontology. New light was shed with the description of Odontochelys, a transitional specimen with a plastron and teeth, but no carapace. Recent studies nested Owenetta (Late Permian), Eunotosaurus (Middle Permian) and Pappochelys (Middle Triassic) as turtle ancestors with teeth, but without a carapace or plastron. A wider gamut phylogenetic analysis of tetrapods nests Owenetta, Eunotosaurus and Pappochelys far from turtles and far apart from each other. Here dual turtle clades arise from a clade of stem turtle pareiasaurs. Bunostegos (Late Permian) and Elginia (Late Permian) give rise to dome/hard-shell turtles with late-surviving Niolamia (Eocene) at that base, inheriting its Baroque horned skull from Elginia. In parallel, Sclerosaurus (Middle Triassic) and Arganaceras (Late Permian) give rise to flat/soft-shell turtles with Odontochelys (Late Triassic) at that base. In all prior phylogenetic analyses taxon exclusion obscured these relationships. The present study also exposes a long-standing error. The traditional squamosal in turtles is here identified as the supratemporal. The actual squamosal remains anterior to the quadrate in all turtles, whether fused to the quadratojugal or not.”

https://pterosaurheresies.wordpress.com/2019/12/04/updated-origin-of-turtles-video-on-youtube/

https://pterosaurheresies.wordpress.com/2016/02/27/the-dual-origin-of-turtles-to-scale/

https://pterosaurheresies.wordpress.com/2016/02/18/turtles-still-not-diapsids/

Another Triassic turtle enters the LRT

This turned out to be a somewhat ‘ho-hum’ event,
but a good opportunity to review turtle origins, still suffering from taxon exclusion and inappropriate taxon inclusion, as determined by the LRT (subset Fig. 4), which tests all contenders.

Sterli, de la Fuente and Rougier 2007 described
“a complete cranial and postcranial anatomy and a phylogenetic analysis of Palaeochersis talampayensis (Rougier, de la Fuente and Arcucci 1995), the oldest turtle from South America, is presented here.”

Figure 1. Palaeochersis skull from Sterli et al. 2007 with colors and new bone labels added.

Figure 2. Palaeochersis overall reconstructed from elements in Sterli, de al Fuente and Rougier 2007. This Triassic turtle nests in the LRT with Proganochelys in figure 3.

Unfortunately
taxon exclusion and bone misinterpretation mar this description, published before the genesis of ReptileEvolution.com and a distinctly different take on turtle origins based on including more taxa. Palaeochersis (Figs. 1, 2; PULR 68; Late Triassic, Norian-Rhaetian) is the third Triassic turtle to enter the LRT after Proganochelys and Proterochersis (Fig. 3).

Figure 3. Proganochelys and Proterochersis, two Traissic turtles.

Sterli, de la Fuente and Rougier 2007 report,
“Palaeochersis talampayensis has primitive character states in the skulllike the presence of lacrimal and supratemporal bones, the presence of a quadrate pocket, a foramen trigemini partially enclosed, middle ear limits partially developed, presence of an interpterygoid vacuity, presence of a cultriform process and a high dotsum sellae. However, Palaeochersis talampayensis also has derived character states like the absence of vomerine teeth, basipterygoid articulation sutured, processus paraoccipitalis tightly articulated to quadrate and squamosal, and cranioquadrate space partially enclosed in a canal. The postcranial description was based on Palaeochersis holotype (PULR 68) and on specimen PULR 69 represented by an almost complete skeleton and a hindfoot, respectively.” 

Figure 4. Subset of the LRT focusing on turles. Palaeochersis is not listed here, but nests with Proganochelys, as shown in the LRT, which has been updated.

According to the LRT
Proganochelys, Proterochersis and Palaeochersis were terminal taxa, leaving no descendants. Instead, according to the large reptile tree (LRT, 1740+ taxa, subset Fig. 4), a Late Permian/Early Triassic sister to Niolamia gave rise to a Late Permian/Early Triassic sister to Meiolania (Fig. 5) gave rise to a Triassic hornless Kallokibotion, otherwise known only from the Latest Cretaceous and later. More derived taxa include Kayentachelys (Early Jurassic) filling in the temporal gap. In turtles the cranial horns are primitive. The loss of cranial horns is a derived trait. That’s something turtle workers have yet to appreciate.

Figure 5. Bunostegos and Elginia at the base of hard shell turtles in the LRT, where late-surviving Niolamia and Meiolania are basal hardshell turtles more primitive than Proganochelys and Palaeochesis. Not all colors here match those in figure 1. This sequence was submitted for publication, but rejected. 

Taxon exclusion mars all prior basal turtle studies.
In the LRT Palaeochersis nests with another Triassic turtle, Proganochelys. Given their overall and detailed morphology, that was expected and comes as no surprise.

Not much else to report here,
other than the identities of cranial bones in hard-shell turtles need to be reconsidered in light of their ancestry from the late-surviving, giant-horned turtles, Niolamia and Meiolania and before that, ancestral pareiasaurs like Elginia and Bunostegos (Fig. 4). These are taxa traditionally omitted from recent turtle ancestry studies. The LRT omits no putative ancestral turtle candidates. Rather the LRT tests them all. And for new readers, note (Fig. 4) soft-shell turtles had their own parallel origin and ancestry from different small horned pareiasaurs.

Longtime readers may notice:
Several of the older turtle skull illustrations (Fig. 5) were recolored to match an emerging standard pattern (e.g. bright green for supratemporals). My fault that I did not do this from the very beginning ten years ago. More work for me, but lookout for different colors in older blogposts.

---

References
Rougier GW, de la Fuente and Arcucci AB 1995. Late Triassic turtles form South America. Science 268:855–858.
Sterli J, de la Fuente MS and Rougier GW 2007. Anatomy and relationships of Palaeochersis talampayensis, a Late Triassic turtle from Argentina. Palaeontographica Abt. A 281:1–61.

wiki/Palaeochersis (Spanish)

Turtle body plans 2020: still not diapsids

Lyson and Bever 2020 once again propose
a diapsid origin for turtles that is not supported by the large reptile tree (LRT, 1719+ taxa, subset Fig. 1) where hardshell and soft-shell turtles arise in parallel from small horned pareiasaurus without temporal fenestrae — and all competing candidates for turtle ancestry are tested.

From the Lyson and Bever abstract:
“The origin of turtles and their uniquely shelled body plan is one of the longest standing problems in vertebrate biology.”

In the LRT this problem has been resolved for several years. Click here for an online paper on the dual origin of turtles from pareiasaurs. Click here for the dual origin of turtles to scale blogpost. Click here for the latest LRT cladogram.

“The unfulfilled need for a hypothesis that both explains the derived nature of turtle anatomy and resolves their unclear phylogenetic position among reptiles largely reflects the absence of a transitional fossil record.”

Not so. In the LRT several overlooked taxa well document a good transitional fossil record. Lyson and Bever omit, ignore and overlook these taxa in favor of several unrelated turtle mimics.

“Recent discoveries have dramatically improved this situation, providing an integrated, time-calibrated model of the morphological, developmental, and ecological transformations responsible for the modern turtle body plan. This evolutionary trajectory was initiated in the Permian (>260 million years ago) when a turtle ancestor with a diapsid skull evolved a novel mechanism for lung ventilation. This key innovation permitted the torso to become apomorphically stiff, most likely as an adaption for digging and a fossorial ecology. The construction of the modern turtle body plan then proceeded over the next 100 million years following a largely stepwise model of osteological innovation.”

Not so. Overlooked taxa known for decades (Elginia (Fig. 2), Sclerosaurus) have been traditionally excluded from turtle origin studies. Some recent discoveries (Eorhynochelys, Pappochelys) nest elsewhere, apart from turtles, as turtle mimics. The LRT tests all known candidates. Lyson and Bever do not. They are still excluding pertinent taxa. Adding more taxa shows that turtles and their ancestors have never been diapsids.

Figure 1. Carbonodraco enters the LRT alongside another recent addition, Kudnu, at the base of the pareiasaurs + turtles.

The most primitive hardshell turtles 
are not the oldest known hardshell turtles. Horned turtle skulls are widely and traditionally considered derived, not primitive. Elginia and Meiolania (Fig. 2) have never been tested together in analysis, and not by Lyson and Bever, despite their obvious similarities and homologies.

Turtle respiration was a big issue for Lyson and Bever.
Earlier we looked at pre-softshell turtle respiration in Sclerosaurus here.

Figure 2. Another gap is filled by nesting E. wuyongae between Bunostegos and Elginia at the base of hard shell turtles in the LRT.

Turtle mimics are out there.
Evidently only a wide gamut phylogenetic analysis, like the LRT, can lump and separate turtle ancestors from turtle mimics without bias and without traditional influences. Lyson and Bever mistakenly accepted several turtle mimics as turtle ancestors, then built a diapsid story around their cherry-picked taxa. Referees and editors also accepted this invalid scenario.

Add taxa
to find and separate real turtle ancestors from turtle mimics.

---

References
Lyson TR and Bever GS 2020. Origin and Evolution of the Turtle Body Plan Annual Review of Ecology, Evolution, and Systematics 51:- (Volume publication date November 2020) Review in Advance first posted online on July 31, 2020. (Changes may still occur before final publication.) https://doi.org/10.1146/annurev-ecolsys-110218-024746

researchgate.net/_The_dual_origin_of_turtles_from_pareiasaurs

Sclerosaurus and the evolution of turtle respiration

Lyson et al. 2014 brought us their view
on the origin of ventilation (= respiration) in turtles using fossils and extant taxa. Similarly, and in the same year, Hirasawa et al. 2014 did the same from a different perspective: turtle embryos.

Unfortunately
neither put their finger on the correct phylogenetic origins of turtles (Fig. 1) due to taxon exclusion. You can’t get a valid phylogenetic solution without a valid phylogeny.

Figure 2. Subset of the LRT focusing on the dual turtle clades (pink) and their ancestors.

Both sets of authors
overlooked/omitted the ancestor taxa of turtles recovered by the large reptile tree (LRT, 1694+ taxa; subset Fig. 1), which tested all current candidates for turtle ancestry. That means both sets of authors stepped into the morass that is convergence.

Figure 2. Sclerosaurus insitu. This turtle ancestor still bas a flexible spine, but the pectoral girdle has migrated anterior to the dorsal ribs. A hypoischiuum is present.

Here,
the LRT (subset Fig. 1) minimizes taxon exclusion due to its wide gamut of included taxa. Here turtles had dual origins from small horned pareiasaurs. Basal to hard-shell turtles, Elginia documents the genesis of cranial traits. Post-crania is poorly known. Basal to soft-shell turtles, Sclerosaurus (Figs. 2–4) documents the genesis of soft-shell turtle traits. These remain (at present) the best clues we have to the genesis of stem hard-shell turtle post-cranial traits. Those are lacking until we go back to the large pareisaur Bunostegos.

Figure 1. Softshell turtle ancestor, Sclerosaurus animated walking in dorsal view. Dorsal armor initially does nothing to prevents lateral undulation here, as shown by the in situ fossil.

Key to the present discussion,
Sclerosaurus had a wide set of dorsal ribs that were not immobilized by the sprinkling of armor over the dorsal vertebrae. The specimen (Fig. 2) is preserved bending far to the left. So it undulated when it walked (Fig. 3). Sclerosaurus lacked a plastron and/or gastralia.

Figure 4. Sclerosaurus walking with an imagined ventral cross-brace, like a turtle plastron. Now Scleromochlus locomotion more closely resembles turtle locomotion. Compare to figure 1.

Immobilzation of the thorax in soft shell turtles
occurs with the genesis of the plastron in Odontochelys (Fig. 5). If we give Sclerosaurus a hypothetical ventral cross brace to stiffen its thorax in the above animated graphic (Fig. 4), it suddenly walks like a turtle (Fig. 4). At first that permits breathing while walking by overcoming Carrier’s constraint. Extant turtles have such a low metabolism that breathing is the last thing they think to do. Sea turtles hold their breath for long periods underwater.

Immobilization of the thorax in Odontochelys
prevented costal ventilation (expanding the ribcage). This is reflected in turtle embryos, which lose intercostal muscles as they develop a rigid shell, according to Hirasawa et al. 2014. Three sets of internal thoracic (hypaxial) muscles take over respiration, expanding to press on the lungs between them or relaxing to initiate inspiration, according to Lyson et al. 2014.

Figure 5. Sister taxa according to Bever et al. Eunotosaurus purportedly nests between Ascerosodontosaurus and the turtles. The large reptile tree, on the other hand, finds that only the turtles are related to each other.

Lyson et al. 2014 
suggested, “the ventilation mechanism of turtles evolved through a division of labour between the ribs and muscles of the trunk in which the abdominal muscles took on the primary ventilatory function, whereas the broadened ribs became the primary means of stabilizing the trunk.” Unfortuantely their ‘early member of the turtle stem lineage’ was the unrelated turtle mimic, Eunotosaurus (Figs. 5, 6). We discussed taxon exclusion errors several times earlier here, here and here.

Figure 6. Subset of the LRT with Martensius added to the base of the Caseasauria + another clade of similar lepidosaurs, all derived from Milleretta. Note the placement of Eunotosaurus with sisters, none of which is close to turtles in the LRT.

Lyson et al. hypothesized,
“an easing of structural constraints through division of function (divergent specialization) between the dorsal ribs and the musculature of the body wall facilitated the evolution of both the novel turtle lung ventilation mechanism and the turtle shell.” This is likely correct, but they used the wrong outgroup taxon, a turtle mimic, rather than a valid stem turtle. Lyson et al. thought the initial thoracic stiffening occurred in the carpace, as it does in Eunotosaurus, which lacks a plastron or more than 5 pairs of slender gastralia not in the radiating pattern of a plastron. Some Eunotosaurus specimens have overlapping ribs. Turtles don’t do this. Mutual side-by-side suturing is the turtle rib pattern and that’s just the beginning of a long list of non-turtle traits found n Eunotosaurus, which nests with Acleistorhinus and other near caseids in the LRT (Fig. 6), all with lateral temporal fenestrae, making them all synapsid mimics.

As you’ll note above,
Sclerosaurus does not have expanded ribs. They begin to expand with Odontochelys (Fig. 5). By contrast, the turtle-mimic, Eunotosaurus, has much more expanded dorsal ribs than those in Odontochelys. That’s the reverse of the order one would expect. The LRT indicates that Lyson et al. should have expanded their taxon list. Sins of omission are also considered sins in paleontology.

Lyson et al. fell prey to a classic error in paleontology
when they ‘Pulled a Larry Martin,‘ listing traits the turtle mimic, Eunotosaurus, shares with turtles. That’s why a good taxonomist saves listing traits until AFTER a comprehensive phylogenetic analysis determines what is related to what and what converges with what.

Hirasawa et al. 2014
attempted to provide ‘answers to the question of the evolutionary origin of the carapace… Along the line of this folding develops a ridge called the carapacial ridge (CR), a turtle‐specific embryonic structure.’ More important to the present discussion is the genesis of the plastron.

A little backstory on Sclerosaurus
Sclerosaurus armatus (Meyer 1859) Middle Triassic ~50 cm in length, was originally considered a procolophonid, then a pareiasaurid, then back and forth again and again, with a complete account in Sues and Reisz (2008) who considered it a procolophonid.

Here, based on data from Sues and Reisz (2008), Sclerosaurus nests between pareiasaurs and basal softshell turtles like Arganaceras, Odontochelys and Trionyx. Their analysis also suffered from taxon exclusion. Sclerosaurus is also a sister to another small horned pareiasaur, Elginia and thus is only slightly more distantly related to Meiolania, the hard-shelled horned basalmost turtle in the LRT.

Overall smaller than other pareiasaurs, Sclerosaurus had a wide, flat body, like the horned lizard, Phrynosoma. The backbone remained quite flexible, as shown by the in situ fossil. Only a sparse sprinking of dermal bones lined the dorsal vertebrae. Note the hypoischium posterior to the ischium and the position of the pectoral girdle anterior to the dorsal ribs, as in Odontochelys.

---

References
Hirasawa T, Pascual‐Anaya J, Kamezaki N, Taniguchi M, Mine K and Kuratani S. 2015. The evolutionary origin of the turtle shell and its dependence on the axial arrest of the embryonic rib cage. J. Exp. Zool. (Mol. Dev. Evol.) 324B:194–207.
Lyson TR et al. (7 co-authors) 2014. Origin of the unique ventilatory apparatus of turtles. Nature Communications 5:5211.
Meyer H von 1859. Sclerosaurus armatus aus dem bunten Sandestein von Rheinfelsen. Palaeontographica 7:35-40.
Sues H-D and Reisz RR 2008. Anatomy and Phylogenetic Relationships of Sclerosaurus armatus (Amniota: Parareptilia) from the Buntsandstein (Triassic) of Europe. Journal of Vertebrate Paleontology 28(4):1031-1042. doi: 10.1671/0272-4634-28.4.1031 online

wiki/Sclerosaurus

Origin of turtle body plan: Schoch and Sues 2019

Schoch and Sues 2019 once again bring us
their invalidated scenario for the origin of turtles (= Odontochelys and Proganochelys) from traditional ancestors Pappochelys, Eunotosaurus and Eorhynochelys (Fig. 1, bottom) that nest elsewhere in the large reptile tree (LRT, 1612+ taxa; subset Fig. 5), which tests these taxa AND pertinent taxa that Schoch and Sues ignore.

Figure 1. Competing turtle origin hypotheses. Bottom: from Schoch and Sues 2019. Top: From the LRT, which tests all taxa from Schoch and Sues, then adds over 1000 more. On the top left are soft-shell turtles and in. On the top right are hardshell turtles and kin. Note the longer tails n club-tailed taxa ignored by Schoch and Sues. Note the use of freehand drawings, which allows bias to creep in and draws us further from the raw data.

Unfortunately,
Schoch and Sues 2019 do not examine and describe the actual ancestors of turtles recovered by the LRT, Stephanospondylus, Bunostegos, Elginia and Sclerosaurus. And because of this they do they do not realize that turtles had a dual ancestry within small, horned pareiasaurs (Fig. 1). Instead, Schoch and Sues cherry-picked their taxa and made up a ‘just-so’ story.

Schoch and Sues are not aware of ‘The Big Picture’
that the first dichotomy splitting all reptiles is the Lepidosauromorpha / Archosauromorpha split. Instead they rely on traditional tree topologies (many of them invalid genomic studies) that do not go back to the origin of reptiles and do not include pertinent taxa. Uncritically they rely on the flawed work of others.

Schoch and Sues do not realize
how rampant convergence is within the Tetrapoda, including the Reptilia. They do not realize that only a wide gamut phylogenetic analysis that recovers a well-resolved tree can correctly nest taxa. Instead they ‘pull a Larry Martin’ and hope the traits and taxa they pick are correct. They are not correct based on testing a wider gamut of taxa in the LRT.

The LRT tells a better ‘just-so’ story
because it does not omit key taxa. And it nests the ancestors hanbd-picked by Schoch and Sues far from turtles on the LRT.

When Pappochelys was added to the LRT,
(Fig. 2) it nested at the base of the Placodontia along with Palatodonta, between Diandongosaurus and Palacrodon and the rest of the placodonts, some of which also had broad ribs and gastralia, all far from turtles. Some placodonts even became turtle mimics.

Figure 2. Pappochelys compared to placodont sister taxa and compared to the Schock and Sues reconstruction, which appears to have several scale bar errors and underestimated the number of dorsal vertebrae. Click to enlarge. So few ribs and vertebrae are known for Pappochelys that their order, size and number could vary from that shown here. Note the ribs of Paraplacodus are also expanded. The number of dorsal vertebrae is unknown and probably more than nine based on sister taxa.

Shoch and Sues 2019
discuss several hypotheses from a hundred years ago or more. They report, “Goodrich (1916) already considered it likely that the absence of temporal openings in turtles represented a secondary condition. His hypothesis has received strong support from the discovery of the Middle Triassic stem-turtle Pappochelys, which has two clearly defined temporal openingson either side of the cranium (Schoch & Sues 2015, 2018).” Seems to fit their scenario, so why not? Testing other candidates (Fig. 1) was apparently never considered.

Not sure why Schoch and Sues keep pursuing this myth.
This is John Ostrom’s lament. Paleontology has always moved at a snail’s pace. Why? Perhaps because workers avoid testing previously published papers and cladograms. It’s not that much work. All they have to do is add taxa, like I do, and the software does the rest.

When the turtle mimics,
Eunotosaurus (Fig. 6) and Eorhinchochelys (Fig. 7), were added to the LRT, they nested together, but with Acleistorhinus and Microleter, taxa also omitted from Schoch and Sues 2019.

Figure 3. Shoch and Sues compared Pappochelys to Odontochelys and Proganochelys, but deleted the more primitive Eunotosaurus. And it’s easy to see why. Eunotosaurus has wider ribs than its two purported successors. That and the LRT tell you its not a turtle, but a turtle mimic. Note the inaccuracy Schoch and Sues applied to their Odontochelys. The version from ReptileEvolution.com appears in frame 2 of this GIF animation.

Figure 4. Eorhynchochelys in situ alongside manus, pes, pectoral and pelvic girdle, plus Eunotosaurus to scale. By convergence Eorhynchochelys resembles Cotylorhychus.

Details:
Schoch and Sues 2019 report, “Specifically, we consider four interconnected issues:

1. what are the closest relatives of turtles among extant reptiles: lepidosaurs (squamates and rhynchocephalians) or archosaurs (crocodylians and birds);
2. what are the closest relatives of turtles among extinct amniotes;
3. how does the bony shell develop in extant turtles; and
4. how did the bony shell evolve in stem-turtles.

Unfortunately
Schoch and Sues do little first hand testing, they rely on genomics and, worst of all, omit taxa that are key to understanding turtle origins (Fig. 8).

Figure 5. Subset of the LRT focusing on turtle origins and unrelated eunotosaurs. Pappochelys nests on a completely different branch of the Reptilia, the Archosauromorpha, close to the base of the Enaliosauria.

Ironic (= by convergence)
that I the updated turtle video came out a few days ago. Here it is again.

I still find it wondrous
that soft-shell turtles and hard-shell turtles evolved so many trait in parallel with roots among the small horned pareiasaurs. It’s all so obvious once you get them together.

---

References
Schoch RR and Sues H-D 2015. A new stem-turtle from the Middle Triassic of Germany and the evolution of the turtle body plan. Nature, 523: 584–587.
Schoch RR and Sues H-D 2018. Osteology of the stem-turtle Pappochelys rosin and the early evolution of the turtle skeleton. Journal of Systematic Palaeontology, 16: 927–965.
Schoch RR and Sues H-D 2019. The origin of the turtle body plan: evidence from fossils and embryos. Palaeontology 2019:1–19.

Read the ResearchGate.net paper on the dual origin of turtles here.

Updated Origin of Turtles video on YouTube

This update
to the original Origin of Turtles YouTube video documents the dual origin of turtles (PDF avaialble on Resarchgate.net) that was not covered in the earlier video, now deleted. This hypothesis was recovered from the large reptile tree (LRT) which now tests 1612 taxa (subset Fig. 2). Other that the turtles themselves, the rest of the included taxa (from birds to fish) are all competing to be the closest outgroup(s) to the turtles. All other candidate taxa, like Eunotosaurus and Pappochelys, are tested in the LRT and featured in the video.

This is a seven-minute video
unfortunately with another eleven minutes of black soundless screen tacked on at the end. Not sure how that happened, but there you are. I would have attempted a repair, but YouTube does not permit identical or near identical videos to be uploaded.

If you want to learn more
about the dual origin of turtles by reading an academic paper on the subject, click this link to ResearchGate.net.

Figure 2. Subset of the large reptile tree (LRT, 1612 taxa) focusing on basal turtles

Abstract from The Dual Origin of Turtles from Pareiasaurs
“The origin of turtles (traditional clade: Testudines) has been a vexing problem in paleontology. New light was shed with the description of Odontochelys, a transitional specimen with a plastron and teeth, but no carapace. Recent studies nested Owenetta (Late Permian), Eunotosaurus (Middle Permian) and Pappochelys (Middle Triassic) as turtle ancestors with teeth, but without a carapace or plastron. A wider gamut phylogenetic analysis of tetrapods nests Owenetta, Eunotosaurus and Pappochelys far from turtles and far apart from each other. Here dual turtle clades arise from a clade of stem turtle pareiasaurs. Bunostegos (Late Permian) and Elginia (Late Permian) give rise to dome/hard-shell turtles with late-surviving Niolamia (Eocene) at that base, inheriting its Baroque horned skull from Elginia. In parallel, Sclerosaurus (Middle Triassic) and Arganaceras (Late Permian) give rise to flat/soft-shell turtles with Odontochelys (Late Triassic) at that base. In all prior phylogenetic analyses taxon exclusion obscured these relationships. The present study also exposes a long-standing error. The traditional squamosal in turtles is here identified as the supratemporal. The actual squamosal remains anterior to the quadrate in all turtles, whether fused to the quadratojugal or not.”

SVP abstracts – Are meiolaniform turtles stem turtles?

Kear et al. 2019 talk about
‘stem’ turtles with skull horns and club tails: the meiolaniforms.

From the abstract:
“Meiolaniforms (Meiolaniformes) are an enigmatic radiation of stem turtles with an exceptionally protracted 100 million-year evolutionary record that spans the mid-Cretaceous (Aptian–Albian) to Holocene. Their fossils have been documented for over 130 years, with the most famous examples being the derived Australasian and southern South American meiolaniids – bizarre horned turtles with massive domed shells and tail clubs that are thought to have been terrestrial and probably herbivorous.”

In the large reptile tree (LRT, 1592 taxa, subset Fig. 2) meiolaniforms (Fig. 1) are not enigmatic. They are basalmost hard-shell turtles derived from similarly-horned Elginia-type small pareiasaurs in parallel with Sclerosaurus-type small pareiasaurs basal to soft-shell turtles.

Figure 2. Another gap is filled by nesting E. wuyongae between Bunostegos and Elginia at the base of hard shell turtles in the LRT.

“Despite a long history of research, the phylogenetic affinities of meiolaniids have proven contentious because of ambiguous character state interpretations and incomplete fossils
representing the most ancient Cretaceous meiolaniform taxa.”

This problem is contentious only because of taxon exclusion. Prior workers have not included analyses of meiolaniforms and Elginia.

“Here, we therefore report the significant discovery of the stratigraphically oldest demonstrable meiolaniform remains, which were excavated from Hauterivian–Barremian high-paleolatitude (around 80°S) deposits of the Eumeralla Formation in Victoria, southeastern Australia. Synchrotron microtomographic imaging of multiple virtually complete skulls and shells provides a wealth of new data, which we combine with the most comprehensive meiolaniform dataset and Bayesian tip-dating to elucidate relationships, divergence timing and paleoecological diversity.”

Did the authors include Elginia, Sclerosaurus, Arganceras and Bunostegos? The abstract does not mention them.

“Our results reveal that meiolaniforms emerged as a discrete Austral Gondwanan lineage,
and basally branching sister group of crown turtles (Testudines) during the Jurassic.”

The LRT invalidated a monophyletic Testudines. Rather soft-shell and hard-shell turtles had separate parallel origins from within the small horned pareisaurs.

Figure 5. Subset of the LRT focusing on turtle origins and unrelated eunotosaurs.

“We additionally recover a novel dichotomy within Meiolaniformes, which split into a unique Early Cretaceous trans-polar radiation incorporating apparently aquatic forms with flattened shells and vascularized bone microstructure, versus the larger-bodied terrestrial meiolaniids that persisted as Paleogene–Neogene relic species isolated in Patagonia and Australasia.”

That’s interesting. The LRT sort of separates the meiolaniform Niolama from the meiolaniform Meiolania + Proterochersis + Proganochelys. The latter taxon also has a club tail. Perhaps more meiolanforms would continue to nest with one or the other.

“Finally, our analyses resolve the paraphyletic stem of crown Testudines, which otherwise includes endemic clades of Jurassic–Cretaceous turtles distributed across the northern Laurasian landmasses. These had diverged from the Southern Hemisphere meiolaniforms by at least the Middle Jurassic, and thus parallel the vicariant biogeography of crown turtles, which likewise diversified globally in response to continental fragmentation and possibly climate.”

Outgroups are key to understanding turtle evolution in the LRT. So is taxon inclusion. Based on the dual origin of turtles from horned small pareiasaurs in the LRT, the list of stem turtles now includes pareiasaurs, if the concept of a monophyletic turtle still stands with a last common ancestor lacking a carapace and plastron within the pareiasaurs.

---

References
Kear BP et al. 2019. Cretaceous polar meiolaniform resolves stem turtle relationships. Journal of Vertebrate Paleontology abstracts.

Evolution of Sea Turtles video has no idea where turtles came from

 A new Ben Thomas video brings us old and invalid view on turtle origins.

I wrote in the comments section:
“Several traditional, but invalid hypotheses in this video. The ability to pull the head inside the shell, whether sideways or straight back is a highly derived character in the hardshell turtle lineage. Sea turtles and their ancestors never could do this. They branched off earlier. Only some soft shell turtles, like the Asian giant soft-shell turtle (Pelochelys) manage to pull half the skull inside a huge mass of scaly flesh by convergence. Eorhynchochelys is a giant eunotosaur. Both are derived from Acleistorhinus not close to turtles. Pappochelys is a basal sauropterygian, again not close to turtles. Sorry, Ben. Eunotosaurus details here:” http://www.reptileevolution.com/eunotosaurus.htm

“Hard shell and soft shell turtles had dual and parallel origins from the small, horned pareiasaurs Elginia and Sclerosaurus. We know turtle origins back to Silurian jawless fish. Cladogram of relationships that tests all published turtle origin candidates here: http://www.reptileevolution.com/reptile-tree.htm

No longer an enigma: Kudnu mackinlayi

I live for discoveries like this one,
which started as a Facebook post of the tiny specimen. This is what the LRT (Fig. 3) was built for.

Benton 1985 wrote:
“Bartholomai (1979) has described Kudnu [QMF8181], a partial snout from the early Triassic of Australia, as a paliguanid. The exact relationships of these forms to each other, and to other early ‘lizard-like’ forms are unclear (Carroll, 1975a, b, 1977; Currie, 1981c: 163-164; Estes, 1983: 12-15). Indeed, the group cannot be defined by any apomorphy, and the genera must be considered separately. As far as can be determined, all of these genera are lepidosauromorphs. Kudnu lacks the lepidosaur character X4 and the squamate character Y 1, but none of the others may be determined. Blomosaurus and Kudnu are classified here as Lepidosauromorpha, incertae sedis.”

Figure 1. Kudnu colorized using DGS and slight restored postcranially, shown 10x natural size at a 72 dpi standard screen resolution. Here’s a taxon basal to Stephanospondylus, pareiasaurs and turtles. Prior workers excluded Stephanospondylus from their studies.

Contrad 2008 wrote:
“Other authors have followed this opinion and have described new ‘‘paliguanids’’, including Blomosaurus (Tatarinov, 1978) and Kudnu (Bartholomai, 1979). Even so, ‘‘Paliguanidae’’is widely regarded as a paraphyletic taxon and, unfortunately, the preservation of specimens constituting the known ‘‘paliguanid’’ genera (including Paliguana, Palaeagama, and Saurosternon) makes it impossible to characterize them except through plesiomorphy (Benton, 1985; Gauthier et al., 1988a; Rieppel, 1994). Thus, their position within Lepidosauromorpha is currently impossible to ascertain with any kind of precision.”

Evans and Jones 2010 wrote:
“Kudnu (Australia, Bartholomai, 1979) and Blomosaurus (Russia, Tatarinov, 1978) are too poorly preserved to interpret with confidence but are probably also procolophonian.”

Figure 2.  Stephanospondylus was considered a type of diadectid, but it nests with turtles and pareiasaurs, all derived from millerettids,.. next to diadectids.

All that being said,
what does the LRT recover? In the large reptile tree (LRT, 1583 taxa, subset Fig. 3) Kudnu nests basal to Stephanospondylus (Fig. 2), a late survivor from deep in the lineage of pareiasaurs + turtles, not far from bolosaurids + diadectids + procolophonids. These clades are derived from Milleretta (Fig. 2) which was  2 to 3x larger.

Due to its small size,
Kudnu can be considered phylogenetically miniaturized, the kind of taxon we often find at the base of many major reptile clades.

Sadly, earlier workers (see above)
were looking at the wrong candidates for sister taxa, excluding the right taxa. This is a problem that is minimized by the LRT due to its large number of taxa over a wide gamut.

Figure 2. Carbonodraco enters the LRT alongside another recent addition, Kudnu, at the base of the pareiasaurs + turtles.Figure 1. Carbonodraco enters the LRT alongside another recent addition, Kudnu, at the base of the pareiasaurs + turtles.

Once again,
you don’t need to see the fossil firsthand in a case like this. What you need is a wide gamut phylogenetic analysis like the LRT, to figure out how an enigma like Kudnu  nests with other reptiles.

If
Kudnu was earlier associated with Stephanospondylus, let me know and I will publish the citation. Otherwise, this is a novel hypothesis of interrelationships that inserts Kudnu without disturbing the rest of the LRT tree topology.

---

References
Bartholomai A 1979. New lizard-like reptiles from the Early Triassic of Queensland. Alcheringa: An Australasian Journal of Palaeontology 3:225–234.
Benton MJ 1985. Classification and phylogeny of the diapsid reptiles. Zoological Journal of the Linnean Society 84:97–164.
Conrad JL 2008. Phylogeny and systematics of Squamata (Reptilia) based on morphology.  Bulletin of the American Museum of Natural History 310: 182pp.
Evans SE and Jones MEH 2010. Chapter 2 The Origin, Early History and Diversification of Lepidosauromorph Reptiles in Bandyopadhyay S (ed.), New Aspects of Mesozoic Biodiversity, Lecture Notes in Earth Sciences 132, DOI 10.1007/978-3-642-10311-7_2 Springer-Verlag Berlin Heidelberg 2010