Figure 1. Pterodactylus with bones color-coded. I think this is Senckenberg n405, Wellnhofer n37. If this wrong let me know. No soft tissue here, but wonderfully articulated. 

Figure 2. Pterodactylus n37, reconstructed. The hatchling and egg are hypothetical. 

I think this is the Senckenberg specimen n. 405 (#37 in the Wellnhofer 1970 catalog). Please let me know if this is not correct. I’d also like to know the scale (length of skull would help.)

This specimen is distinct from the other Pterodactylus specimens already featured in ReptileEvolution.com and nests about midway between the more primitive forms and the more derived forms. Like derived taxa, the scapula is no longer than the coracoid and pedal digit 5 has only one digit. Unlike the other Pterodactylus specimens, no pedal digits are disc-like (shorter than their width).

Earlier we looked at Diadectes, noting that it nests deep inside the plant-eating side of the Reptilia, the new Lepidosauromorpha.

Let’s take a look at that skull again. 
Wiki reports, “Among its primitive features, Diadectes has a large otic notch (a feature found in all labyrinthodonts, but not in reptiles) with an ossified tympanum.” Other reptiles with a large otic notch include several close relatives of Diadectes, including a sister taxon, Procolophon (Fig. 1). The resemblance is not just superficial, yet Wiki reports, “Procolophon was a genus of lizard-like procolophonid reptiles.” Why was Procolophon considered a reptile and Diadectes an amphibian? It can’t be the notch. It’s the same on both. Procolophon was simply smaller diadectid that lived later in time. Take a look a the various Diadectes skulls  on the Wiki page and you’ll see that the otic notch is bigger on some, smaller on others. It’s not homologous with the similar structure in amphibians. Seymouria retains an intertemporal bone and has palatal fangs. Diadectes does not.

Figure 1. In the large reptile tree Procolophon nests with Diadectes, and both share a large otic notch, a trait Wiki says makes Diadectes an amphibian.

The Otic Notch
The otic notch simply redeveloped by convergence in diadectids and procolphonids, yet one got labeled a reptile and one an amphibian. I don’t know why. Both Procolophon (Owen 1876) and Diadectes (Cope 1878a, b) were first described long ago. Perhaps this is some sort of tradition from a time when we didn’t know very much about prehistory. If anyone has original literature, I’d like to see it.

Ancestors in the Large Reptile Tree
Orobates has a small otic notch and nests primitive to the diadectids and procolophonids. Tseajaia, Solenodonsaurus and the chroniosuchids all have an otic notch and all are considered by Wiki to be amphibians, but here nest in the Reptilia. All these taxa and the diadectids + procolphonids have Concordia in their pedigree. It has virtually no otic notch, but you can see how it could have begun here. Funny thing is, Wiki reports, Concordia is close to the origin of the captorhinid reptiles, and it is too in the large reptile tree. These sorts of problems emphasize the importance of adding lots of taxa to basal reptile studies. The more you add, the more Diadectes nests with Procolophon deep inside the Reptilia.

Herbivorous
Wiki reports, “Diadectes was one of the very first herbivorous tetrapods.” One look at the chronological reptile tree indicates that two other herbivores, Cephalerpeton and Concordia preceded Diadectes chronologically. So do Orobates and Stephanospondylus. Captorhinids are likewise herbivores, but they are found in younger rocks despite their more primitive nesting.

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

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

References
Cope ED 1878a. Descriptions of extinct Batrachia and Reptilia from the Permian formation of Texas. Proceedings of the American Philosophical Society 17:505-530.
Cope ED 1878b. A new Diadectes. The American Naturalist 12:565.
Owen R 1876. Descriptive and Illustrated Catalogue of the Fossil Reptilia of South Africa in the Collection of the British Museum. London, British Museum (Natural History).

Figure 1. Plate for Eocypselus rowei with soft tissue preservation.

Both swifts and hummingbirds have smaller feet and legs than those of Eocypselus rowei. Because of this, along with their extraordinary flying abillities, swifts and hummingbirds forego walking for the most part. In contrast, Eocypselus rowei appears to have been a good walker with longer metatarsals and legs.

Figure 2. Eocipselus rowei counterplate distorted to match plate. Evidently the plate and counter plate were not taken from the exact same viewpoint.

Eocypselus rowei had a stout humerus (Fig. 6) but not so stout as either a swift or hummingbird, both of which were relatively 2/3 the length and 1/3 deeper (Fig. 6). Likewise in the swift and hummingbird the radius/ulna is about 2/3 of the length in Eocypselus rowei. The manus of the swift and hummingbird is much longer than the combined length of the ulna and humerus (Fig. 4), but not so in the more generalized and primitive Eocypselus rowei.

Figure 3. Tracing of Eocypselus, identifying bones by color. DGS used here to trace elements.

Lead author Daniel Ksepka reported, “This fossil bird represents the closest we’ve gotten to the point where swifts and hummingbirds went their separate ways.”

Figure 4. Reconstruction of Eocypselus rowei. The pelvis is preserved in ventral view, so is difficult to ascertain in lateral view, but it probably looked very much like that of most other similar birds. Also shown is Apus (illustration from Eyton 1867), the modern common swift, in which the hand bones exceed the humerus + ulna in length.

Eocypselus vincenti (Harrison 1984, Mayr 2010, Fig. 5) is a congeneric specimen from the Early Eocene of Europe.

Figure 5. Eocypselus vincenti, a related species from Europe from Mayr (2010). Apparently the manus is slightly larger and the tibia is slightly smaller here than in E. rowei. Note the lack of scapulocoracoid fusion here. 

Mayr (2010) also found Eocypselus vincenti to be related to swifts and hummingbirds. Harrison (1984) originally named Eocypelus. Others have also found and described this genus.

Figure 6. The evolution of the humerus in Eocypselus, swifts and hummingbirds, rearranged and colored from Mayr 2003. In both swifts and hummingbirds the humerus becomes increasingly robust and in both a new process develops (1, 3) that originates in Eocypselus.

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

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

Reference
Eyton TC 1867. Osteologia avium; Or, a Sketch of the Osteology of Birds / II. : Wellington, London.
Harrison CJO 1984. A revision of the fossil swifts (Vertebrata, Aves, suborder Apodi), with descriptions of three new genera and two new species. Mededelingen van de Werkgroep voor Tertiaire en Kwartaire Geologie 21:157–177.
Ksepka DT, Clarke JA, Nesbitt SJ, Kulp FB and Grande L. 2013. Fossil evidence of wing shape in a stem relative of swifts and hummingbirds (Aves, Pan-Apodiformes). Proceedings of the Royal Society B: Biological Sciences 280 (1761): 20130580. doi:10.1098/rspb.2013.0580. Supplementary materials here.
Mayr G 2003. Phylogeny of early Tertiary swifts and hummingbirds (Aves: Apodiformes). The Auk 120(1):145–151, 2003. online
Mayr G 2009. Paleogene Fossil Birds (online) Springer.
Mayr G 2010. Reappraisal of Eocypselus—a stem group apodiform from the early Eocene of Northern Europe. Palaeobiodiversity and Palaeoenvironments 90(4): 395-403.

Discover Magazine online

wiki/Eocypselus
Fossil hummingbirds online pdf

Figure 1. Click to enlarge. Pterodaustro specimen MIC-V263 two wing tips. The labeled one clearly shows a wing ungual (m4.5, duplicated and rotated in the small box). The other m4.4 (in the large box) does not show an ungual, despite the slight break that matches the other ungual length.  Rather the ungual was lost off the edge of the matrix (see Fig. 2, lower edge). Typically to universally unguals are crushed onto their wider face, showing their essential claw–like character.

I think it’s unusual and atypical that the tip of m4.4, nearest the alleged ungual, is expanded so much. Not sure exactly what’s going on there. Let’s just throw this up as a possibility, despite its apparent clarity. Also unexpected is the texture on the broken portion of m4.4 that makes it look more like a drill bit.

Nice to see someone else is also tracing photographs
I have been vilified for tracing photographs using Adobe Photoshop (computer software). Here, it’s clear that Codorniú et al. (2013) traced the photograph of MIC-V263 (Fig. 2), the first step in DGS, because there is a perfect correspondence between drawing and photo. That doesn’t happen very often otherwise.

Figure 2. Click to enlarge. Pterodaustro DGS tracing by Codorniú et al. (2013). There’s a perfect correspondence of elements. See, DGS is not such a bad thing after all. Even though they traced m4.5 (the wing ungual) they did not identify it as such, overlooking its significance.

Interestingly
The gastroliths are all located between the ilia. This must have happened when the crop was shifted posteriorly during taphonomy.

Wing digit 5
I see some probable m4.5 elements (Fig. 3), but they are shifted from their typical orientation.

Figure 3. Pterodaustro possible wing digit 5, rotated medially. Ungual in lavender. Other phalanges in baby blue. Metacarpal in green. Carpal in magenta. Looks like all the other examples I have ever found. Worth keeping a lookout for. 

What else?
Pedal digit 5 appears to have one phalanx, but the end of p5.1 is a hinge joint and the next bone is barely visible beneath it.

Manual digit 3 is seen for the first time as longer than m digit 2. This is distinct from Ctenochasma in which the digits 2 and 3 are subequal. Manual 3.2 in Pterodaustro is not a disc. This was corrected in the reconstructions of both the adult (Fig. 4) and embryo. The femur also has a longer neck than I originally thought.

Figure 4. Pterodaustro adult with manual digit 3 repaired.

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

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

References
Codorniú L, Chiappe LM and Cid FD 2013. First occurrence of stomach stones in pterosaurs, Journal of Vertebrate Paleontology, 33:3, 647-654.

Conway has the eye of a true artist. His work is simply beautiful. He also brings new insight into familiar and not so familiar specimens. His choice of colors, point-of-view and lighting are unique and more than satisfying. His work invites close inspection and admiration. His work evokes mood and involvement.

Here’s a selection from his homepage.

Figure 1. The art of John Conway. Click to go to enlarge and go to his website.

Sure I have the usual rant/quibbles
about his Rhamphorhynchus (he followed the invalidated Sordes cruropatagium model of Sharov/Bakhurina/Unwin), but those are easily overlooked when seduced by his talents for portraying it. In any case, Conway illustrated this falsified hypothesis more clearly than anyone else ever and, in doing so, answered the persistent question: “Was the cloaca above or below the ‘cruropatagium’?” [Conway indicates it was below, evidently, making sex a wee bit more difficult, but excrement did not stain the membrane].

Earlier I also made notes on his Pteranodon proportions.

Don’t miss his Anhanguera cutaway. It’s a classic. Be sure to run your mouse over the “Skeleton :: Musculature :: Pulmonary :: External“ caption to see all four images. A truly amazing illustration.

The origin of turtles
is one of the most hotly debated topics in paleontology. Unique among living amniotes, turtles have a carapace, plastron and a shoulder girdle within the rib cage. Some DNA studies point to an archosaur link. Other studies link turtles to lizards. Only the large reptile tree looked at over 335 possible nesting sites for turtles and came up with one.

Dr. Michael S. Y. Lee (1993) provided the best published morphological report on turtle origins to date. This paper precedes the discovery of Odontochelys (Li et al. 2008), overlooks Stephanospondylus (Geinitz and Deichmuller 1882) and nests turtles with pareiasaurs like Deltavjatia (Hartmann-Weinberg 1937).

Figure 1. The pareiasaur Deltavjatia identifying turtle traits: A2 – Foramen palatinum medially located (similar to the suborbital fenestra). A8 – Supraoccipital forms a long, high, narrow and median ridge sutured to the skull roof along its entire length. A12 – Scapula with acromion process on the anterior margin. A13 – Humerus with ectepicondylar foramen. B1 – (Fig. 1) Twenty or fewer presacral vertebrae. B2 – Tall and narrow scapula (4x higher than wide). B3 – Shoulder glenoid not screw-shaped, but bipartite. B4 – Scapula oriented anterodorsally, not horizontally. B8 – Thick dermal armor over the dorsal region.

In Lee (1993) pareiasaurs were found to share 16 derived traits with turtles. These traits are identified with an “A“.

Cranial traits synapomorphies:
A1 – (Fig. 2) Choana (internal nares) located far medially.
A2 – (Figs. 1,2) Foramen palatinum medially located (similar to the suborbital fenestra).
A3 – (Fig. 2) Massive horizontal paroccipital process sutured to squamosal.
A4 – (Fig. 2) Long lateral flange of the exoccipital on the posterior face of the paroccipital process.
A5 – (Fig. 2) Basisphenoid and basioccipital ossified together.
A6 – (Fig. 2) Ossified medial wall of prootic.
A7 – (Fig. 2) Transverse flange of pterygoid reduced and forwardly directed.
A8 – (Fig. 1`) Supraoccipital forms a long, high, narrow and median ridge sutured to the skull roof along its entire length.
A9 - The entire palate is raised well above the ventral margin of the maxilla.

Figure 2. More turtle traits in pareiasaurs. A1 – Choana located medially. A2 -Foramen palatinum medially located. A3 – Massive horizontal paroccipital process sutured to squamosal. A4 – Long lateral flange of the exoccipital on the posterior face of the paroccipital process. A5 – Basisphenoid and basioccipital ossified together. A6 – Ossified medial wall of prootic. A7 – Transverse flange of pterygoid reduced and forwardly directed. A9 – The entire palate is raised well above the ventral margin of the maxilla.

Postcranial Trait Synapomorphies
A10 – (Fig. 3) Prominent lateral projections on at least the first 14 caudal vertebrae.
A11 – (Fig. 3) Chevrons not wedged between adjacent centra.
A12 – (Figs. 1, 3) Scapula with acromion process on the anterior margin
A13 – (Fig. 1) Humerus with ectepicondylar foramen.
A14 – (Fig. 3) Femur with a major trochanteron the posterior margin.
A15 – (Fig. 3) Reduced pedal digit 5.
A16 – Prominent dorsal buttress, V-shaped in ventral view, overhanging the acetabulum.

Figure 3. Postcranial turtle traits in pareiasaurs. A10 – Prominent lateral projections on at least the first 14 caudal vertebrae. A11 – Chevrons not wedged between adjacent centra. A12 – Scapula with acromion process on the anterior margin. A14 – Femur with a major trochanteron the posterior margin. A15 – Reduced pedal digit 5.

Sclerosaurus
Nine more traits are shared by Proganochelys, pareiasaurs and Sclerosaurus, the smaller, flatter, pareiasaur sister. These are identified with a “B” by Lee (1993).

B1 – (Fig. 1) Twenty or fewer presacral vertebrae.
B2 – (Figs. 1, 3) Tall and narrow scapula (4x higher than wide).
B3 – (Figs. 1, 3) Shoulder glenoid not screw-shaped, but bipartite.
B4 – (Figs, 1-3) Scapula oriented anterodorsally, not horizontally.
B5 – Reduced manual phalangeal formula (23332)
B6 – (Fig. 3) Astragalus and calcaneum fused
B7 – (Fig. 3) Reduced pedal phalangeal formula (23343)
B8 – (Fig. 1) Thick dermal armor over the dorsal region.
B9 – Loss of gastralia.

The large reptile tree found Sclerosaurus to be a derived pareiasaur, not closer to turtles. Chronologically Stephanospondylus preceded turtles and Sclerosaurus by 70 million years. Stephanspondylus preceded pareiasaurus by 35 million years, plenty of time for these radiations to occur. Look for primitive turtles in the mid to late Permian, concurrent with pareiasaurs.

But wait, there’s more…
The large reptile tree used only a few of the above traits yet to likewise nest turtles with pareiasaurs and Sclerosaurus. Stephanospondylus does not preserve any palate, tail, manus femur, pes or armor data.

The scapula question
Lee notes that pareiasaurs and Sclerosaurus possess 5 cervicals and 14-15 dorsals for a total of 19 to 20. Turtles possess 8 cervicals and 10 dorsals, meaning that 3 turtle cervicals are former dorsals. This change was accompanied by a posterior shift of the pectoral girdle (Watson 1914) that is recapitulated during turtle ontogeny (embryogenesis).

All known pareiasaurs are too pareiasaur-y to be ancestral to turtles
*Stephanospondylus is a key taxon linking diadectids to pareiasaurs and turtles that avoids being to “pareiasaur-y.” No known archosaur shares so many turtle traits. No known sauropterygian comes close either. Out of 335+ taxa, Stephanospondylus remains the best candidate I’ve found. But, sans that taxon, turtles would nest just outside the Pareiasauria.

Hats off to Dr. Lee for doing a great job.

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

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

References
Geinitz HB and Deichmüller JV 1882. Die Saurier der unteren Dyas von Sachsen. Paleontographica, N. F. 9:1-46.
Hartmann-Weinberg AP 1933. Evolution der Pareiasauriden: Trudy Palaeontological institute Academe Nauk, SSSR, 1933, n. 3, p. 1-66.
Lee MSY 1993. The Origin of the Turtle Body Plan: Bridging a Famous Morphological Gap. Science 264:1716-17-1719.
Li C, Wu X-C, Rieppel O, Wang L-T, Zhao L-J 2008. An ancestral turtle from the Late Triassic of southwestern China. Nature 456: 497-501.
Romer AS 1925. Permian amphibian and reptilian remains described as Stephanospondylus. Journal of Geololgy 33: 447-463.
Stappenbeck R 1905. Uber Stephanospondylus n. g. und Phanerosaurus H. v. Meyer: Zeitschrift der Deutschen Geologischen Gesellschaft, v. 57, p. 380-437.
Watson DMS 1914. Eunotosaurus africanus Seeley and the ancestors of the Chelonia. Proceedings of the Zoological Society of London 11:1011.

Palaos discussion

Traditional thinking follows the paradigm
that the unfused scapulocoracoid (s/c) in pterosaurs demonstrates immaturity. I tested this in a phylogenetic analysis. Turns out the patterns are not ontogenetic, but clearly phylogenetic. Scapulocoracoid fusion is on again, off again in patterns that are not the random pattern one would expect if ontogenetic in nature.

Figure 1. Click to enlarge. Pterosaur family tree (May 2013) highlighting scapulocoracoid fusion in pterosaurs (bright green) and lack of fusion (bright blue). Other taxa do not preserve the s/c. If ontogenetic we would expect a more scattered, randomized pattern. That’s not the case here as fusion patterns follow phylogeny, not maturity. Some taxa here do not preserve the scapula and coracoid. Not listed here, but related to Cearadactylus, Barbosania does not fuse the s/c. Some taxa have complete fusion. Others retain a line of fusion. Among the higher ornithocheiridae there is the greatest randomness in fusion.

Pterodaustro is known from embryos to fully mature individuals
Codornú et al. (2013) report on 300+ individual specimens from a single bone bed: ”Interestingly, proxies for full skeletal maturation are thus far present only in isolated elements (i.e., all complete or semicomplete specimens belong to osteologically immature individuals). These proxies include the complete fusion (lack of any sutural evidence) between the extensor tendon process and the shaft of the first wing phalanx, the complete fusion between the tibia and the proximal tarsals, and the fused distal secondary ossification centers of the humerus.” Note they did not report fusion of the scapula and coracoid. That’s because Pterodaustro nests in a clade (Fig. 1) that does not fuse the scapulocoracoid.

So what’s the pattern?
Basal pterosaurs do not have a fused scapulocoracoid. Dimorphodon may have a fused s/c. Campyognathoides and basal Dorygnathus fuse the s/c. Basal Rhamphorhynchus specimens are smaller and lack fusion. Derived Rhamphorhynchus regain fusion. Dorygnathid pre-azhdarchids beginning with tiny TM 10341 lose fusion. Large azhdarchids regain fusion. No ctenochasmatid or dorygnathid pre-ctenochasmatid fuse the scapulocoracoid. Jianchangnathus and all subsequent scaphognathids lose fusion. Basal ornithocheirds, no matter how large their wings are do not fuse the s/c. Certain, but not all derived ornithocheirds regain fusion. On another branch of scaphognathids, certain germanodactylids regain fusion. Shenzhoupterids and basal tapejarids lose fusion. Derived tapejarids, the big ones, regain fusion. (Does anyone have a good dsungaripterid scapulocoracoid? I haven’t seen one yet.) Germanodactylids including Pteranodon have fusion (not sure about basal taxa because so many are known just by skulls), but eopteranodontids and nyctosaurs lack scapulocoracoid fusion.

A little pterosaur referred to Eudimorphodon, BsP 1994 has a fused s/c. Arthurdactylus a much larger, longer winged ornithocheirid, does nto fuse the s/c. So size is not the issue.

All known pterosaur embryos come from clades that do not fuse the scapulocoracoid. However, the  juvenile Pteranodon has a fused s/c.

Addendum
Once a clade began to fuse the s/c, then lack of fusion generally accompanied phylogenetic size reductions. Among azhdarchids, only Quetzalcoatlus fuses the s/c. This includes a smaller Pteranodon YPM2525 which may also represent a size reduction shown here.

Among the derived ornithocheirds you do get a more randomized on-off-on-off pattern.

So there you have it. All results subject to change with injections of new valid data.

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

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

References
Maisano JA 2002. The potential utility of postnatal skeletal developmental patterns in squamate phylogenetics. Journal of Vertebrate Paleontology 22:82A.
Maisano JA 2002. Terminal fusions of skeletal elements as indicators of maturity in squamates. Journal of Vertebrae Paleontology 22: 268–275.

Figure 1. Kuehneosaurid skulls from Palaegama to Coelurosauravus and Mecistotrachelos, and to Lanthanolania, Pamelina, Kuehneosaurus, Icarosaurus and Xianglong. Some of these taxa were not previously recognized as kuehneosaurids or their ancestors.

Gliding struts and lateral extradermal membranes probably first appeared as decorations because the skull of Coelurosauravus (Fig. 1) is also distinctly decorated with a squamosal/supratemporal frill. Mecistotrachelos (Fig. 1) kept its frill. The others did not. The frill-less taxa also lose the supratemporal, a bone that makes up the back part of the frill. Palaegama has no dermal struts. Lanthanolania is known by a skull only. Pamelina vertebrae (Fig. 2) do not have the long fused transverse processes of Kuehneosaurus.

Widely considered to glide with hyper-elongated ribs, the lineage of kuehneosaurids indicates that those ribs were actually ossified dermal filaments/bones (as seen in Coelurosauravus, Fig. 1). The ribs shrank (Fig. 2) and disappeared and in their place grew elongated transverse processes to act as anchors for the gliding struts. This happened by convergence twice, once in Mecistotrachelos and again in the lineage of kuehneosaurids without a frill. You can see the transformation in Kuehneosaurus and Pamelina (Fig. 2).

Figure 2. Above, sample vertebrae from Pamelina. Below, the more derived Kuehneosaurus. True ribs are shown in yellow. Dermal struts are in blue. Note the lack of fused transverse processes on the dorsal vertebrae in Pamelina. The elongated caudal transverse processes indicate the presence of a large caudofemoral muscle, which would have been much smaller in Kuehneosaurus.

A paper by Evans (2009)
described Pamelina (Fig. 1), an early Triassic kuehneosaurid added a third genus to her list of kuehneosaurs. Note the lack of fused transverse processes on the dorsal vertebrae in Pamelina. The elongated caudal transverse processes indicate the presence of a large caudofemoral muscle, which would have been much smaller in Kuehneosaurus.

Other members
Earlier we also added Xianglong (which is not a lizard), and Lanthanolania (which is not a younginoid) along with Coelurosauravus (Fig.1), which everyone else thinks developed rib membranes by convergence.

Family Tree
Evans (2009) produced an interesting family tree of the Reptilia. Except for turtles it splits reptiles into two main lineages, one that includes lepidosaurs and one that includes archosaurs, like the large reptile tree does. Of course the tree by Evans assumes the outgroups and basal taxa include synapsids, which nest in the archosaur half of the large reptile tree.

Figure 2. Reptile tree according to Evans 2009, that is very much in line with the large reptile tree, except for the nesting of turtles (probably due to shelled placodonts) near Sauropterygians. Blue = the new epidosauromorphs. Yellow = the new archosauromorphs.

The current state of phylogenetic thinking
Evans (2009) reports, “The Neodiapsida of Benton (1985) encompasses a wide range of diapsid lineages, most of which can be assigned to either Archosauromorpha or Lepidosauromorpha (Gauthier et al. 1988). Archosauromorpha encompasses a large and successful crown clade (Archosauria) and a series of distinctive stem lineages (e.g., protorosaurs, tanystropheiids, Prolacerta, Rhynchosauria, Trilophosauria, Evans 1988; Gauthier et al. 1988; Müller 2002, 2004; Modesto and Sues 2004). Crown−group Lepidosauria (Rhynchocephalia and Squamata) also constitutes a large and diverse group but, leaving aside the issue of testudine or sauropterygian affinities (e.g., Rieppel and de Braga 1996; de Braga and Rieppel 1997; Rieppel and Reisz 1999; Müller 2002, 2004; Hill 2005).”

By contrast the large reptile tree found those listed members of Evans’ “Neodiapsida” to be diphyletic and found the lepidosauromorpha also include tanystropheids, Rhynchosauria and Trilophosauria along with turtles.

We’re working for consensus, but first others have to expand their inclusion set gamut and avoid suprageneric taxa.

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

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

References
Evans SE 2009. An early kuehneosaurid reptile from the early Triassic of Poland. Palaeotologia Polonica 65: 145-178.

The problem is turtles (Figs. 1,2) are different from all living animals and distinct from most prehistoric ones. Their roots are very deep. Earlier we looked at taxa that looked like turtles, but were not turtles. Here we’ll expand that list.

What are turtles?
Phylogenetically what we’re looking for is the reptile most like a turtle that is not yet a turtle. We know of several turtle-like reptiles (see below), but none share a turtle’s basic anatomy. They all developed their protective shells via convergence.

Genes
A recent (May 15, 2012) online story here reports from Crawford et al. (2012), “Scientists lift lid on turtle evolution. Anatomy and fossil studies of turtles and their reptilian relatives generally place the shelled creatures in the family of lepidosaurs — snakes, lizards and tuataras (rare lizard-like animals). Genetic studies, however, say they have more in common with crocodiles and birds.”

Of course,
if turtled did descend from archosaurs and their kin, then we have to look for the closest relatives of turtles in the new Archosauromorpha. Trouble is, given the widest gamut of possibilities yet offered, turtles nest in the other lineage, of lizards and their kin, not with archosaurs. Other respected genetic studies, like Lyson et al. (2012), report turtles are genetically closer to lizards.

So how do we solve this problem?
We can’t. All three sides have their proof.

Of course, someday you’ll have to find a non-turtle that looks like a turtle, and that’s phylogeny and morphology. So, here we’re going to look at seven distinct types of reptiles (including mammals) that had a turtle-like morphology. We’ll start with turtles themselves.

Figure 1. Proganochelys from the Late Triassic. Formerly the most primitive turtle.

Figure 2. Odontochelys, also from the Late Triassic, the new most primitive turtle. It has teeth. The carapace is missing. Lost or not yet developed has not been determined yet.

Figure 3. Stephanospondylus from Romer (1925). According to the results of the large reptile tree, this is the most turtle-like non-turtle yet discovered. And yet, among all these taxa, it’s the only one without a shell or scutes (that I know of). Chronologically Stephanospondylus precedes the Odontochelys by 70 million years, plenty of time to iron out the differences and plenty of time to find a transitional taxon or ten, seven million years apart from each other someday. No carapace or plastron was preserved with Stephanospondylus.

1. Turtles themselves
Derived from Stephanospondylus, turtles like Odontochelys and Proganochelys have a plastron, but only Proganochelys has a carapace. Thus, Odontochelys was analogous to a “soft-shelled” turtle, but not directly related. Turtles have fewer than ten dorsal ribs and have no temporal fenestrae. Stephanospondylus is a little-studied diadectomorph close to pareiasaurs that happens to share more traits with turtles than any other studied taxon. Unfortunately, it is incompletely known and crushed. We studied Stephanospondylus earlier here. I hope other paleontologists will begin to consider this long neglected taxon in their turtle studies, at least to test the large reptile tree results. In order to do so, they will also have to recognize the reptile traits of diadectomorphs. Unfortunately at present diadectomorphs are widely considered to be non-amniotes close to basal amniotes.

Figure 2. Cyamodus, a sharp-snouted shelled placodont.

2. Cyamodontids
Derived from Palatodonta, cyamodontids like Cyamodus and Placochelys have a carapace and a second smaller one over the hips. Huge upper temperal fenestra and flat-pebble-like teeth differentiate this placoderm from turtles.

Figure 3. Henodus, a broad-snout shelled placodont

3. Henodus
Derived from a sister to Placodus, Henodus is another placodont with an independently evolved wrap-around carapace and tiny legs projecting out of anterior and posterior openings. Tiny upper temporal fenestra and a broad rostrum differentiated this taxon from turtles and cyamodontids.

Figure 5. Sinosaurosphargis. Click for more information. Not sure about the anterior extent of the maxilla, here shown two ways. Ribs and gastralia like these are not known in Omphalosaurus, which has more plesiomorphic and typical looking ribs and gastralia.

4. Sinosaurosphargis
Derived from Claudiosaurus, Sinosaurosphargis and Largocephalosaurus have a carapace covering dozens of wide flat ribs and gastralia. The nostrils are located midway between the long snout tip and orbit.

Figure 2. Eunotosaurus, a milleretid not related to turtles.

5. Eunotosaurus
Derived from Acleistorhinus, Eunotosaurus has nine expanded ribs, like turtles, but no true carapace or plastron. Lateral temporal fenestra are present. The tail is exceptionally long.

Figure 7. Ankylosaurus, dorsal view. This early image was made prior to the discovery of tail clubs in ankylosaurs, but reflects the distribution of osteoderms over the bak.

6. Ankylosaurs
Derived from basal ornithischian dinosaurs like Scelidosaurus, ankylosaurs like Ankylosaurus, had a carapace made of osteoderms.

Figure 8 Stagonlepis, an aetosaur derived from rauisuchians

7. Aetosaurs
Derived from odd rauisuchians like Ticinosuchus, aetosaurs like Stagonlepis (Fig. 8) also had a carapace made of osteoderms.

8. Glyptodonts and Armadillos
Derived from armadillo-like ancestors, glyptodonts like Glyptodon, had a carapace made of hexagonal scutes, otherwise known as bony skin scales called osteoderms.

I’m not including Jaxtasuchus, an armored protorosaur, which we looked at earlier. It could be number nine, but it’s getting longer and thus less turtle-like.

So, turtle-like anatomies are fairly common in the prehistoric past. Now, not so much, but turtles themselves display a wide variety of types and sizes and niches. Most paleontologists understand that these taxa are all distinct from turtles. However, when some of these taxa are included in phylogenetic analyses without also including Stephanospondylus, pareiasaurs and diadectids, then turtles tend to be attracted to these shelled wonders and thus skew the results toward placodonts or eunotosaurs (Lyson et al. 2010).

Just want to give the rightful ancestors their due. The strength of the large reptile lies in its ability to differentiate the true ancestors of turtles from the convergent pretenders.

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

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

References
Crawford NG, Faircloth BC, McCormack JE, Brumfield RT, Winker and Glenn TC 2012. More than 1000 ultraconserved elements provide evidence that turtles are the sister group of archosaurs. Biology Letters, 2012; DOI: 10.1098/rsbl.2012.0331.
Lyson TR, Bever GS, Bhullar B-AS, Joyce WG and Gauthier JA 2010. Transitional fossils and the origin of turtles. Biology Letters 6 (6): 830–833. doi:10.1098/rsbl.2010.0371.
Lyson TR, Sperling EA, Heimberg AM, GauthierJA, King BL, and Peterson KJ 2011. MicroRNAs support a turtle + lizard clade. Biol Lett 2011 : rsbl.2011.0477v1-rsbl20110477.abstract - online news story.
Rieppel O and DeBraga M. 1996. Turtles as diapsid reptiles. Nature 384 (6608): 453–5. doi:10.1038/384453a0.

In the prehistoric past there was a third lepidosaur clade, the Tritosauria, that nests just outside the Squamata in the large reptile tree. Tritosaurs became extinct at the end of the Cretaceous with the end of the Pterosauria.

Another extinct clade, one more basal to Lepidosaurs, gave rise to the so-called rib-gliders, typified by the kuehneosaurids. Altogether these taxa are considered the Lepidosauriformes with Paliguana (Fig. 1) at the base.

I thought it might be interesting to focus on the basal taxa from each of these clades. Perhaps not surprisingly, they’re quite similar to each other, yet each one gave rise to a variety of derived forms, from pterosaurs to snakes to mosasaurs to tanystropheids to rhynchosaurs to chameleons.

Let’s start at the beginning (more or less) of the Lepidosauriformes
Owenettids, like Owenetta, were basal to Lepidosauriformes according to the large reptile tree. They had a lateral termporal fenestra, but no lower temporal bar and no upper temporal fenestra. Owenettids appear to have been ground dwellers. Owenetta had wide gracile ribs giving it a wide flat body.

Figure 1. Basal lepidosauriformes and lepidosaurs. These morphologies are more similar to each other than to highly derived taxa in each of these distinct clades. 

Lepidosauriformes
Paliguana (basal lepidosauriformes) is known by its skull only. It is the earliest/most basal taxon in this lineage with upper temporal fenestrae.

Long-legged Saurosternon (not shown in Fig. 1) leads a splinter lineage that became arboreal and ultimately produced so-called rib-gliders, Coelurosauravus and the Kuehneosauridae. This taxon and all subsequent forms did not have a wide torso.

Lepidosauria (Sphenodontia + Tritosauria + Squamata)
Gephyrosaurus was basal to rhychocephalians (= sphenodontians) including trilophosaurs and rhynchosaurs. The scapula was more robust and fused to the coracoid. The pelvis had a thyroid fenestra.

Tritosauria + Squamata
Dalinghosaurus was basal to the Tritosauria, a newly identified lepidosaur clade that ultimately gave rise to drepanosaurs, tanystropheids and pterosaurs. Lacertulus (late Permian) and Homoeosaurus were also basal members. Note the relatively longer, stronger hind limbs. Some of these became slow-moving arboreal forms (drepanosaurids). Others remained agile and sometimes bipedal terrestrial forms (fenestrasaurs leading to pterosaurs). Some of these later became aquatic, long-necked and gigantic (tanystropheids).

Squamata
Iguana is a member of the Iguania and a basal squamate. All four limbs are robust with large unguals.  Here we find a shorter neck and slender pubis/enlarged thyroid fenestra. Some were terrestrial, others arboreal. One  (marine iguanas) ventured into the sea.

Scleroglossa
Liushusaurus (Fig. 1) is a basal member of the Scleroglossa and also a basal squamate. The forelimb was as large as the hind limb. Basal forms were terrestrial venturing into arboreal (Gekko). Some of these lost their legs (Lialis). Others lost their legs and became fossorial (burrowers like amphisbaenids). Another clade became fossorial (Heloderma) and also lost their legs (Cylindrophis and kin). Still another clade became marine. Some became giants (mosasaurs). Others from this clade also lost their legs (snakes and kin).

In Summary
Lepidosaurs crawled, burrowed, climbed, glided, flew and swam. Some were giants. Most were not. Some were extremely tiny as adults (some gekkos and some pterosaurs). Most were cold-blooded. Pterosaurs were covered with fibers and thus were probably warm-blooded. Some laid eggs and ignored them. Others retained eggs until just before hatching. Others bore live young. Some developed a lower temporal bar. Some lost both of their temporal bars. Some sealed up their lower temporal opening.

Each of these clades started off looking pretty much alike (Fig. 1) having descended from a common ancestor and then diversifying via evolution, each according to its own niche and environs.

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

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