Turtle origins: Pappochelys STILL not the best candidate

Schoch and Sues 2017
bring us more details about Pappochelys, and pull a ‘Larry Martin‘ or two to force fit this taxon into a false narrative: the origin of turtles story. What little they report and show is indeed intriguing. What more they don’t report and show invalidates their hypothesis. A wider gamut phylogenetic analysis has the final say.

As a reminder,
many paleontologists try to find one, two or a dozen traits that look like they link one taxon to a clade, but avoid testing those hypotheses in a wide gamut phylogenetic analysis, like the large reptile tree (LRT, 1048 taxa). This technique of force-fitting and ignoring other candidate sisters never turns out well. It’s not pseudoscience, but it does remind one of early attempts at flying that did not include sufficient power, rudders, ailerons and horizontal stabilizers. Those attempts were all doomed to crash.

A wide gamut phylogenetic analysis
remains the only tool that always delivers a correct tree topology because  taxon exclusion is minimized. The LRT worked with Diandongosuchus. It worked with Lagerpeton. It worked with Chilesaurus. It worked with turtles, whales and seals. And it worked with pterosaurs. The LRT works!

Let’s just make this short and painful
Schoch and Sues ignored:

  1. the sister of Pappochelys in the LRT, Palatodonta
  2. other proximal relatives of Pappochelys in the LRT, Diandongosaurus, Anarosaurus, Palacrodon and Majiashanosaurus
  3. the sister to hard shell turtles in the LRT, Elginia
  4. the sister to soft shell turtles in the LRT, Sclerosaurus
  5. basalmost hard shell turtles in the LRT, Niolamia and Meiolania.
  6. the proximal relatives of Eunotosaurus in the LRT, Acleistorhinus, Delorhynchus, Australothyris and Feeserpeton.
Figure 1. Shoch and Sues cladogram of turtle origins. Look at that loss of resolution!

Figure 1. Shoch and Sues cladogram of turtle origins. Look at that loss of resolution! Gliding kuehneosaurs nest between aquatic taxa? Really? Add about 300 taxa and let’s see if this tree resolves itself. 

Schoch and Sues employed only 29 taxa
many of which were suprageneric, compared to 1048 specimens in the LRT. Schoch and Sues lament, “the currently available data fail to support any of the three more specific hypotheses for the diapsid origins of turtles (sister group to Sauria, Lepidosauria or Archosauria, respectively). We found no support for earlier hypotheses of parareptilian relationships for turtles hypothesized by Laurin & Reisz (1997) and Lee (1997), respectively, nor for the hypothesis that captorhinid eureptiles were most closely related to turtles (Gaffney & McKenna 1979; Gauthier et al. 1988).” Schoch and Sues published a cladogram (Fig. 1)  in which the following taxa could not be resolved:

  1. Acerosodontosaurus (swimming diapsid)
  2. Kuehneosauridae (gliding lepidosauriforms)
  3. Claudiosaurus (swimming diapsid)
  4. ‘Pantestudines’ = Eunotosaurus, Pappochelys, Odontochelys, Proganochelys (turtles and turtle mimics)
  5. Trilophosaurus + Rhynchosauria + Prolacerta + Archosauriformes (a paraphyletic mix)
  6. Squamata + Rhynchocephalia (terrestrial lepidosaurs)
  7. Placodus + Sinosaurosphargis + Eosauropterygia (swimming enaliosaurs)

In other words
Schoch and Sues have no idea how these taxa are related to each other. Their data fails to lump and separate 29 taxa completely. They report, “[Papppochelys] shares various derived features with the early Late Triassic stem-turtle Odontochelys, such as T-shaped ribs, a short trunk, and features of the girdles and limbs.” See what I mean about pulling a ‘Larry Martin’? They’re trying to save their hypothesis by listing a few to many traits. Unfortunately Schoch and Sues do not have the data that documents this suite is unique to Pappochelys and turtles. Actually these traits are found elsewhere within the Reptilia and sometimes several times by convergence.

Figure 1. Pappochelys comes to us from several specimens, all incomplete and all disarticulated. These are the pieces of the skull we will use in Photoshop to rebuild the skull. Schock and Sues made a freehand cartoon, a practice that needs to be discouraged.

Figure 2. Pappochelys comes to us from several specimens, all incomplete and all disarticulated. These are the pieces of the skull we will use in Photoshop to rebuild the skull. Schock and Sues made a freehand cartoon, a practice that needs to be discouraged. They had the nasals backwards and the lacrimal upside down and labeled a prefrontal. The failed to recognized the quadratojugal. And they changed the squamosal. The postorbital looks to be so fragile that the orbit might instead have been confluent with the lateral temporal fenestra.

Freehand reconstructions
Shoch and Sues created their reconstructions not by tracing bones, but freehand. That never turns out well. They created cartoon bones and modified them to be what they wanted them to be when they could have used Photoshop and real data.

Figure 2. 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 3. In dorsal view 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. Since Pappochelys is know from 4 or more scattered and incomplete specimens, we really don’t know how many dorsal ribs it had.

Why didn’t they show Eunotosaurus
(in Fig. 3)? This turtle mimic has wider and more extensive dorsal ribs. That could be one reason. We’re all looking for a gradual accumulation of traits and Eunotosaurus, one of many turtle mimics, does not provide the primitive state.

Figure 6. 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.

Figure 4. From two years ago. Pappochelys compared to placodont sister taxa and compared to the Schoch and Sues reconstruction, which appears to have several scale bar errors and underestimated the number of dorsal vertebrae. So few ribs and vertebrae are known for Pappochelys that their order, size and number could vary from that shown here. 

The ‘Probably’ weasel word
Pappochelys is not known from any complete or articulated fossils. Even so Shoch and Sues report, “The vertebral column of Pappochelys comprises probably eight cervical, probably nine dorsal, two sacral, and more than 24 caudal vertebrae.” This is wishful thinking… They should have said ‘unknown’ not ‘probably’.

Dredging up false data to support a diapsid relationship
Schoch and Sues reference Bever et al. (2015) when they show a Eunotosaurus juvenile purportedly lacking a supratemporal and in its place, an upper temporal fenestra. Earlier that ‘missing’ supratemporal was identified as a nearby bump on the cranium of the juvenile.

Gastralia
Turtle ancestors in the LRT have no gastralia. So the origin of the plastron is still not known. According to Schoch and Sues, “The gastralia of Pappochelys are unique in their structure and arrangement.” Unfortunately Palatodonta is only known from cranial remains.    All other proximal relatives in the LRT have slender gastralia, not broad like those in Pappochelys. Some Pappochelys gastralia are laterally bifurcated, similar to the plastron elements in Odontochelys. That’s intriguing, but ultimately yet another Larry Martin trait. What we’re looking for is maximum parsimony, a larger number of traits shared by sister taxa and proximal relatives than in any other taxa.

Scapula
The Pappochochelys scapula is dorsally small and slender, like those of other placodonts and basal enaliosaurs. Shoch and Sues compared it to the basal turtle scapula, which is relatively much larger. Comparable pectoral elements are documented in the outgroups Bunostegos and Sclerosaurus, but these were ignored by Shoch and Sues. We don’t know of any post-crania for the hard shell turtle sister, Elginia, which might or might not have had a Meiolania-like carapace.

Shoch and Sues made some great observations,
but they kept their blinders on with regard to other candidates. A wide gamut analysis really is the only way to figure out how taxa are related to one another. Hand-picking traits and cherry-picking a small number of taxa is not the way to understand turtle origins. However, once relationships are established and all purported candidates are nested in a large gamut analysis, THEN it’s great to describe and compare how various parts of verified sister taxa evolved.

The LRT
nests turtles with pareiasaurs. Hardshell turtles arise from the mini-pareiasaur Elginia to Niolamia. Softshell turtles arise from the mini-pareiasaur Sclerosaurus to Odontochelys. Pappochelys nests with Palatodonta at the base of the Placodontia.

References
Bever GS, Lyson TR, Field DJ and Bhular B-A S 2015. Evolutionary origin of the turtle skull. Nature published online Sept 02. 2015.
Schoch RR and Sues H-D 2017.
Osteology of the Middle Triassic stem-turtle
Pappochelys rosinae and the early evolution of the turtle skeleton. Journal of Systematic Palaeontology DOI: 10.1080/14772019.2017.1354936

Leave a Reply

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

WordPress.com Logo

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

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s

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