Loss of resolution in cladograms

A cladogram
is a graphic image (a diagram) typically generated from software, and typically based on a list of taxa and a list of characters, each one scored for each taxon. The cladogram is supposed to model actual evolutionary relationships among organisms. Ideally every cladogram will be fully resolved with every taxon sufficiently different from its sisters to merit its own node or branch. This is the basis for lumping and splitting. In practice loss of resolution occurs when sisters are not sufficiently different from one another. This can happen due to one or more of several problems:

  1. Three taxa are in reality too closely related. As an example, you may have input several specimens of the same genus and species, like “Tyrannosaurus rex,” without having characters in your characters list that could taxonomically separate them. Two identical taxa nest together, no problem. Three identical taxa produce loss of resolution. We saw something like this earlier in pterosaurs were two Rhamphorhynchus specimens had the same score and were deemed to be an adult and juvenile of the same species – but the third sister taxon (of the same putative species!) did not have the same score.
  2. Two sister taxa share no characters in common despite being closely related. This occurs most often when a skull-only taxon nests as a sister to a skull-less taxon, but it could also occur with other combinations of missing parts.
  3. One taxon is in a ‘by default’ nesting. It should not be in your taxon list because in reality it is not closely related to any other taxon in your taxon list. For instance, when one attempts to nest the pterosaurs “Pteranodon” or “Dimorphodon” with generic and specific members of the Archosauria or Archosauriformes, or when one attempts to nest “Homo sapiens” with generic and specific members of the Ichthyosauria. No telling what can happen in those instances, but anyone is able to try.
  4. Too few characters #1. If you have only 5 or 50 characters, there may not be a large enough list of traits to split your taxa apart. Statistically the list becomes large enough for 98% certainty at around 150 characters and becomes incrementally better with every character added thereafter. The large reptile tree uses 228 characters, many with more than two options, and has complete resolution (except for skull-less and skull-only taxa nesting as sisters (#2 above) with 546 taxa. Thus in practice there is no 3:1 character:taxon ratio as you may have learned about in theory.
  5. Too few characters #2. If your sister specimens are only represented by a few bones or parts of bones then two sister taxa may not resolve.
  6. Too few characters #3. Legless burrowing tetrapods appear to converge in their remaining traits so you better have a sufficient number of traits to lump and split the various clades. Otherwise the legless clades tend to be attracted to one another.
  7. Too few taxa. This goes back to #3 above because by throwing in one ‘by default’ taxon, like “Vancleavea” into an unrelated clade, like “Archosauriformes,” without also including the verified sisters of “Vancleavea,” like “Helveticosaurus” and “Askeptosaurus,” might produce loss of resolution because “Vancleavea” shares so few traits with any archosauriforms and the addition of the other thalattosaurs will clarify relationships. You may not have loss of resolution, but by adding taxa you’ll eliminate these ‘by default’ taxa.
  8. Using suprageneric taxa. If you can pick traits from one partial specimen AND another partial specimen to get enough traits to fill a suprageneric character list for a single taxon, then you’ve created a chimaera that can only lead to trouble. Even if your two taxa are incomplete, that’s better than creating a single cherry-picked chimaera taxon.
  9. Mistakes in scoring. Since humans are scoring all characters, mistakes can happen and they can affect nestings. Mistakes are often due to: 1) trying to maintain a paradigm or tradition; 2) too much leeway or opinion possible in a choice of possible scoring options; 3) inadvertent transpositions of data; 4) typos; 5) relying on the veracity of prior scorings, etc. Double and triple check your work and the work of others when you find loss of resolution. Errors are easy to make.
  10. Loss of resolution can occur at several levels: 1) in a simple heuristic search you need only one character score to lump and split sister taxa from your list of several dozen to several hundred traits; 2) in a bootstrap/jacknife search you need at least three character scores to lump and split taxa to raise your bootstrap score over 50%.

In my manuscripts,
when I report that my trees are fully resolved, that never seems to impress the referees (or they don’t believe it). Perhaps that is so because so many accepted manuscripts have loss of resolution at several nodes for many of the above reasons. That is not acceptable in most cases (exception: skull-only taxa will continue to occasionally nest with skull-less taxa).

We know better now.
We now have a large gamut “umbrella” study that continues to recover relationships within the Reptilia as it continues to increase in size. This large study provides a basis for smaller, more focused studies. When the old unverified traditions and paradigms have been replaced with verified models and relationships, then we’ll all have more confidence in recovered trees.

Nesting turtles with pterosaurs redux 2011-2015

For those who don’t read the ‘Letters to the Editor’,
a recent comment on sister taxa inspired me to revisit the old experiment that nested pterosaurs and turtles together as a result of taxon exclusion, which you can review here.

By default nestings
can be interesting and silly. The point behind nesting pterosaurs with turtles back then was to examine the folly behind nesting pterosaurs with archosaurs — only possible due to a similar taxon exclusion that’s been going on for at least fifteen years now, following the publication of a phylogenetic analysis that nested pterosaurs with fenestrasaurs (Peters 2000) and has been ignored ever since.

Back in the day (July 2011) with 360 taxa,
when all other taxa were removed from the lepidosauromorph side of the large reptile treeProganochelys, the turtle, nested with MPUM6009, the pterosaur at the base of the Sauropterygia. That’s bizarre, but interesting and hopefully enlightening by analogy to the achosaur-link question.

Today,with 508 taxa,
and deleting all other lepidosauromorphs, the pterosaur now nests between Cathayornis and Struthio, the ostrich, + Gallus, the chicken. The turtle now nests with the frogs, between Doleserpeton and Gerobatrachus + Rana.

Hmm. Let’s fix that.
Let’s delete the amphibians and add the basal lizard Huehuecuetzpalli and guess what happens?

The three lepidosauromorphs:
the turtle, the lizard and the pterosaur, all nest together again in their own clade at the base of the Sauropterygia… in other words, nowhere near dinos, pre-dinos, parasuchians, Lagerpeton or Marasuchus. Delete Huehuecuetzpalli and Proganochelys nests with the turtle-like placodont, Henodus, as you might imagine, while the basal pterosaur bounces back to the birds. So one taxon in-between the turtle and pterosaur were needed this time to glue them together in a single clade and to trump the attraction of other candidate sisters.

Bottom line:
by including more and more taxa the large reptile tree provides more and more nesting sites, and thus the large reptile tree minimizes unwanted ‘by default’ nestings. Up to now other workers have been relying an tradition and paradigm for their taxon lists, and many of those traditions have been tested (and falsified) at reptileevolution.com. When workers base their smaller, more focused studies on a larger umbrella study, they will have greater success and greater confidence that their cladogram is a good one = with no ‘by default’ nestings.

Eoraptor, Panphagia and Pampadromaeus: how closely are they related?

Lumping and splitting
is something paleontologists do with the various specimens they find as they assign them names and nodes in the family tree of life. Jack Horner recently made news for lumping several pachycephalosaur genera together as distinct ontogenetic growth stages of the same genus and species. He did the same with Triceratops, which changed its appearance rather drastically while reaching maturity.

Today
let’s look at three closely related specimens, Panphagia protos Martinez and Alcobar (2009) , Eoraptor lunensis (Sereno et al. 1993, 2014) and Pampadromaeus barberenai (Cabriera et al. 2011, Fig. 1). These three nest between basalmost theropods and basalmost phytodinosaurs (sauropodomorphs + ornithischians) in the large reptile tree. Others consider them basal or stem sauropodomorphs, but only because basal ornithischians are not included in their analyses.

Figure 1. Eoraptor, Pampadromaeus and Panphagia. Three coeval South American dinosaurs between Theropoda and Phytodinosauria. Are they congeneric?

Figure 1. Eoraptor, Pampadromaeus and Panphagia. Three coeval South American dinosaurs between Theropoda and Phytodinosauria. Are they congeneric? Or are they distinct enough to be considered separate genera? To my eye, they appear to be morphs of a single genus.

The number of bones preserved
with each specimen varies, so all the bones cannot be compared with one another. What is preserved, however, appears to be more closely matched than many other specimens sharing the same generic name, like Pteranodon and Rhamphorhynchus. (Evidently dino-workers are splitters and ptero-workers are lumpers as, until recently, they preferred not to provide new names for distinct specimens, some of which were improperly considered juveniles of distinctly different larger specimens).

These three proto-phytodinosaurs (Fig. 1) are obviously similar.  Sereno, et al. (2014) often refers to “the closely related Pampadromaeus and Panphagia” when writing about Eoraptor.

What are the differences? 

Using the 228 characters of the large reptile tree is not enough to split and lump these three specimens. Only these three traits split them and cause loss of resolution.

  1. The anterior nasal is wider in Panphagia.
  2. The mandible tip does not descend in Pampadromaeus.
  3. The tibia is shorter than 2x the ilium length in Pampadromaeus

More resolution might come from adding taxa and more complete taxa, unless these three are indeed congeneric. Almost certainly there are obscure, but important differences not covered by the list of 228 rather obvious traits.

Enter Martinez et al. (2012).
Their strict consensus tree (51 taxa, 378 characters) was also unable to resolve relationships among these three and several other taxa. Their reduced consensus tree (eliminating poor specimens) nested them in ascending order: Phanphagia > Eoraptor > Pampadromaeus separated by single decay indices.

The position of Panphagia as basal to other sauropodomorphs is supported by nine unambiguous synapomorphies:

  1. pterygoid wing of the quadrate extending for more than 70% of the total quadrate length;
  2. presence of postparietal fenestra between supra occipital and parietals;
  3. supraoccipital wider than high;
  4. coarse serrations of the teeth angled upwards at 45◦;
  5. absence of a  postzygodiapophyseal lamina in cervical vertebrae 4–8;
  6. weakly developed laminae in the neural arches of cervical vertebrae 4–8;
  7. minimum width of the scapula less than 20% of its length;
  8. posterior end of the fibular condyle of the tibia anterior to the posterior margin of proximal articular surface;
  9. and strongly laterally curved iliac blade in dorsal view.

The more derived position of Eoraptor is supported by three unambiguous  synapomorphies:

  1. subtriangular cross-section of the ischial midshaft;
  2. supraacetabular crest of the ilium contacting the distal end of pubic peduncle;
  3. and sub triangular distal end of the ischium.

Pampadromaeus and more derived sauropodomorphs share four unambiguous synapomorphies:

  1. squamosal bordering the laterotemporal fenestra for more than 50% of its depth (62:0)
  2. length of the base of proximal caudal neural spines greater than half the length of the neural arch;
  3. transverse width of the distal humerus greater than 33 of its length;
  4. and length of the pubic peduncle of the ilium greater than twice the anteroposterior width of its distal end.

So the question remains,
are these several distinctions sufficient to split these three specimens? Are they indeed distinct genera? Or are they all three species of Eoraptor?  (Eoraptor is the earliest of these three to be named.)

References
Cabreira SF, Schultz CL, Bittencourt JS, Soares MB, Fortier DC, Silva LR and Langer MC 2011. New stem-sauropodomorph (Dinosauria, Saurischia) from the Triassic of Brazil. Naturwissenschaften (advance online publication) DOI: 10.1007/s00114-011-0858-0
Martínez RN and Alcober OA 2009. A basal sauropodomorph (Dinosauria: Saurischia) from the Ischigualasto Formation (Triassic, Carnian) and the early evolution of Sauropodomorpha (pdf). PLoS ONE 4 (2): 1–12. doi:10.1371/journal.pone.0004397. PMC 2635939. PMID 19209223. online article
Martínez RN , Apaldeti C and Abelin  D 2012. Basal sauropodomorphs from the Ischigualasto Format ion, Journal of Vertebrate Paleontology, 32:sup1, 51-69.
Sereno PC, Forster CA, Rogers RR and Moneta AM 1993. Primitive dinosaur skeleton form Argentina and the early evolution of the Dinosauria. Nature 361, 64-66.
Sereno PC, Martínez RN and Alcober OA 2013. Osteology of Eoraptor lunensis (Dinosauria, Sauropodomorpha). Basal sauropodomorphs and the vertebrate fossil record of the Ischigualasto Formation (Late Triassic: Carnian-Norian) of Argentina. Journal of Vertebrate Paleontology Memoir 12: 83-179 DOI:10.1080/02724634.2013.820113

wiki/Eoraptor
wiki/Panphagia
wiki/Pampadromaeus

Adelosaurus: transitional between Claudiosaurus and Atopodentatus

Another roadkill fossil gets nested
Adelosaurus huxleyi (Hancock and Howse 1870, Evans 1988, Figs. 1, 2) was originally considered to be a small protorosaur distinct from Protorosaurus (Watson 1914). It is the poster child for roadkill fossils, spread out on a plate and lacking a skull. Evans (1988) reported, “In the absence of the skull and ankle, classification remains tentative. The skeleton seems immature.” Otherwise there’s not much out there on Adelosaurus.

Phylogenetic analysis brings new insight
as Adelosaurus now nests between Claudiosaurus germaini (Carroll 1981, Late Permian) and the odd new kid on the block (not known to anyone before last year), Atopodentataus unicus (Cheng et al. 2014, early Middle Triassic).

Figure 1. Adelosaurus, a genuine roadkill fossil from the Late Permian together with a reconstruction of same. Note the dorsal expansion of the clavicle, and the robust scapulocoracoid.

Figure 1. Adelosaurus, a genuine roadkill fossil from the Late Permian together with a reconstruction of same. Note the dorsal expansion of the clavicle, and the robust scapulocoracoid.

The first step in understanding any roadkill, even if the only data is a crude drawing, is to rearrange the parts into its in vivo position, realizing that mistakes can be corrected later, following Steve Jobs guidelines.

Figure 2. Adelosaurus (middle) nests between Claudiosaurus (top) and Atopodentatus (bottom), all at the base of the Enaliosauria.

Figure 2. Adelosaurus (middle) nests between Claudiosaurus (top) and Atopodentatus (bottom), all at the base of the Enaliosauria. Click to enlarge. 

Adelosaurus has a pectoral girdle, manus and carpus nearly identical to those of Claudiosaurus (Fig. 2). Digit 4 of the manus lacks two phalanges, unlike Claudiosaurus, but that’s an odd trait found  in Atopodentatus. The clavicle of Adelosaurus is interesting. Evidently it doesn’t have a broad medial portion. No sister taxa do either. That’s the dorsal portion, nearly identical to those odd clavicles in an Atopodentatus sister, Largocephalosaurus. Atopodentatus has standard straight clavicles, or so it appears. Adelosaurus is also the most basal taxon with a curved humerus, an enaliosaur trait.

Ultimately Adelosaurus is the ‘small plain brown sparrow’ from which all the odd and wonderful variations within the Enaliosauria appear. So it needs to appear in all future phylogenetic studies that employ member clades.

References
Cheng L, Chen XH,Shang QH and Wu XC 2014. A new marine reptile from the Triassic of China, with a highly specialized feeding adaptation. Natur
Carroll RL 1981. Plesiosaur ancestors from the Upper Permian of Madagascar. Philosophical Transactions of the Royal Society London B 293: 315-383.
Currie PJ 1981. Hovasaurus bolei, an aquatic eosuchian from the Upper Permian of Madagascar. Palaeontologica Africana, 24: 99-163.
Evans 1988. The Upper Permian reptile Adelosaurus from Durham. Palaeontology 31(4): 957-964. online pdf
Hancock A and Howse R 1870. On Protorosaurus speneri von Meyer, and a new species, Protorosaurus huxleyi, from the Marl Slate of Middridge, Durham. Quarterly Journal of the geological Society of London 26, 565-572.
Watson DMS 1914. Broomia perplexa gen. et. sp. nov., a fossil reptile from South Africa. Proceedings of the Zoological Society, London 1914:995-1010

 

wiki/Claudiosaurus

Heleosuchus – the enigma has nested in the Rhynchocephalia

Figure 1. Heleosuchus, a former enigma, nests in the middle of the Rhynchocelphalia, between Planocephalosaurus and Sphenodon.

Figure 1. Heleosuchus, a former enigma, nests in the middle of the Rhynchocelphalia, between Planocephalosaurus and Sphenodon. Here is Heleosuchus in situ and Planocephalosaurus restored to scale. Click to enlarge.

Heleosuchus (Fig. 1) has been an enigma since first described by Owen 1876. Several heavy-hitters in paleontology (Broom 1913, Evans 1984, Carroll 1987) have taken a whack at it without resolving its relations.

According to Wikipedia, “It was originally described as a species of Saurosternon, but was later recognized as a separate taxon by R. Broom. Heleosuchus is suggested as being either an early diapsid reptile, not closely related to other lineages, or as being an aberrant and primitive lepidosauromorph. Heleosuchus shares the hooked fifth metatarsal found in some other diapsids, such as primitive turtles (Odontochelys), lepidosauromorphs, and archosauromorphs, but it also resembles ‘younginiform’-grade diapsids in its gross morphology.  Heleosuchus may also share a thyroid fenestra with these higher diapsid reptiles as well, but the identity of this feature is disputed.”

Based on tracings by Carroll (1987) the large reptile tree (not updated yet) Heleosuchus nested between Planocephalosaurus (Fig. 1) and the clade of Sphenodon and Kallimodon in the middle of the Rhynchocephalia. What was identified as a scapula must be a portion of the interclavicle instead.

However, even Carroll was not sure of the identification of several elements. Unfortunately it appears as though the last time someone published on Heleosuchus was prior to the advent of computer-assisted phylogenetic analysis. Carroll notes, “if a thyroid fenestra is present and the fifth metatarsal is hooked, Heleosuchus would definitely represent a lineage distinct from the younginoids. These features are present in Late Triassic sphenodontids and Jurassic lizards, but they are also present in other groups. In conclusion, the characters that are preserved point to a position near the base of the lepidosauromorph assemblage, possibly close to the younginoids but perhaps representing a distinct lineage.”

What appears to be bothering Carroll is the early appearance of Heleosuchus in the Late Permian of South Africa relative to the lepidosaurs known to him at the time. That early appearance doesn’t bother the large reptile tree, which nests several other Permian contemporaries just as high if not higher in the reptile family tree.

References
Broom R 1913. A revision of the reptiles of the Karroo. Annals of the South African Museum 7: 361–366.
Carroll RL 1987. Heleosuchus: an enigmatic diapsid reptile from the Late Permian or Early Triassic of southern Africa”. Canadian Journal of Earth Sciences24: 664–667.
Evans SE 1984. The anatomy of the Permian reptile Heleosuchus griesbachi. Neues Jahrbuch für Geologie und Paläontologie, Monatshefte, 12: 717-727.
Owen R 1876. Descriptive and illustrated catalogue of the fossil Reptilia of South Africa in the collection of the British Museum. Trustees of the British Museum (Natural History), London, UK.

wiki/Heleosuchus

Adding Homo sapiens to the large reptile tree

Not sure why I didn’t think to do this earlier.
I added Homo sapiens to the large reptile tree (still not updated) and ran a phylogenetic analysis to see where we nest. To no one’s surprise Homo (Fig. 1) nested with the cynodont Procynosuchus among the Therapsida (no other mammals are yet entered).

Figure 1. Homo sapiens alongside sister taxa Australopithecus and Ardipithecus (both in gray).

Figure 1. Homo sapiens alongside sister taxa Australopithecus and Ardipithecus (both in gray). Click to learn more.

Entering characters for Homo and Procynosuchus
was more than enlightening as so many traits were shared between the two. You should try it sometime!

Figure 2. Procynosuchus, a basal cynodont therapsid synapsid sister to humans in the large reptile tree (prior to the addition of advanced cynodonts including mammals).

Figure 2. Procynosuchus, a basal cynodont therapsid synapsid sister to humans in the large reptile tree (prior to the addition of advanced cynodonts including mammals). Click to learn more.

What happens when taxa are excluded? (How deep can we go?)
The contribution of cynodont traits to the story of human evolution was more powerful than I thought. I was surprised at one happened when I took one step further.

Deleting only Procynosuchus
results in Homo nesting results in Homo nesting between Dibamus and Tamaulipasaurus (Fig. 2) two burrowing skinkomorph squamates. My guess is the fusion/loss of so many skull bones, the brevity of the rostrum, the great depth of the coronoid process of the dentary and the complete lack of postcranial characters for the two burrowing taxa are attracting the taxon Homo with similar skull traits.

With the present character and taxon list, these skinkomorphs nest closer to humans than Biarmosuchus and more basal synapsids, like Dimetrodon. They’re just not human enough.

Biarmosuchus, the most basal therapsid.

Figure 3. Biarmosuchus, the most basal therapsid and not a cynodont. Despite nesting as a basal therapsid, its traits do not attract the taxon Homo more than others do. 

You think THAT’S ridiculous. Let’s take the next step…

Figure 2. Tamaulipasaurus nests with Homo sapiens when the basal cynodont, Procynosuchus, is excluded.

Figure 2. Tamaulipasaurus nests with Homo sapiens when the basal cynodont, Procynosuchus, is excluded. The fusion of skull bones, the short rostrum, and the large coronoid process of the dentary are traits shared with humans.

Deleting all the skinkomorph squamates
results in Homo nesting as a turtle/pareiasaur ancestor. Here the short face, anterior nares, tall pelvis and loss of manual and pedal phalanges appear to attract Homo to turtles like Proganochelys.

Proganochelys. Formerly the most primitive turtle.

Figure 2. Proganochelys. Formerly the most primitive turtle. Click to learn more. 

See what happens with taxon exclusion?
Strange bedfellows can result. So many current problems and enigmas in paleontology can be readily settled with a large enough family tree.

In the same light, I challenge paleontologists
to add thalattosaurs to Vancleavea studies… to add fenestrasaurs to pterosaur studies… to add mesosaurs to ichthyosaur studies… and to add millerettids to caseasaur studies. There’s no harm in doing so, and we all might learn something.

 

Restoring Scoloparia as a procolophonid AND as a pareiasaur

Today I have a quandary…
Is Scoloparia a procolophonid or a pareiasaur? I’ve looked at it both ways (Figs. 1, 2). It nests both ways (depending on the restoration), and at least one way is wrong.

This problem highlights more basic problems
found within the Procolophonidae, some of which nest in the large reptile tree (still not updated)  with diadectids (Procolophon and kin, Fig. 1), with pareiasaurs (Sclerosaurus) and the rest nest as pre-Lepidosauriformes (Owenetta and kin). Conventionally procolophonids are considered parareptiles. Cisneros lists Nyctiphruretus as the outgroup and owenettids as basal taxa within the Procolophonidae. The large reptile tree replicated that outgroup only for the owenettids.

Scoloparia glyphanodon (Sues and Baird 1998) is currently represented by several specimens, three of which are figured, colorized and restored here (Figs. 1, 2). All three differ in size. Comparable skulls differ in morphology. This has been attributed to ontogeny.

Figure 1. Scoloparia restored here as a procolophonid together with other procolophonids.

Figure 1. Scoloparia restored here as a procolophonid together with other procolophonids. Click to enlarge. The large YPM mandible is a definite procolophonid. The small 82.1 specimen is a definite procolophonid. The holotype is the big question mark.

Clearly the referred specimens
(the dentary and the small 82.1 specimen) are procolophonids. Only seven blunt and rotated teeth in a mandible that tips down anteriorly along with gigantic orbits mark these taxa as procolophonids. They compare well with other procolorphonids.

Figure 2. Scoloparia restored as a pareiasaur close to Elginia along with several other pareiasaurs for comparison. Sclerosaurus typically nests as a procolophonid, but even with the removal of all skull traits, it nests as a small pareiasaur.

Figure 2. Scoloparia restored as a pareiasaur close to Elginia along with several other pareiasaurs for comparison. Sclerosaurus typically nests as a procolophonid, but even with the removal of all skull traits, it nests as a small pareiasaur. The new restoration reidentifies several bones. Note the convergence with the procolophonids in figure 1.

The problem is in the large holotype
The 83.1 specimen holotype of Scoloparia was preserved without a skull roof or palate, so the nasals, frontals and parietals are restored here.

Originally
the size and morphological differences were attributed to the juvenile status of the smaller specimen. H. Sues wrote to me, “Both specimens have the same peculiar ‘cheek’ teeth, which are unlike those of any other procolophonid.” 

I think what Dr. Sues means is shown below in figure 3. The teeth of the referred specimen attributed to Scoloparia have multiple cusps, unlike most procolophonids, but approaching the serrated morphology of pareiasaurs. The convergences are mounting!! And now you see why this is a quandary!

Figure 4. Teeth compared. Elginia, Scolaparia (referred), Leptopleuron and Diadectes.

Figure 3. Teeth compared. Elginia, Scolaparia (referred), Leptopleuron and Diadectes, a stem procolophonid. Oddly the very procolophonid Scoloparia (referred specimen) does have peculiar teeth for a procolophonid. They are serrated somewhat like those in the pareiasaur, Elginia.

I have asked to see images of the teeth for the Scoloparia holotype. No reply yet.

The mystery of the holotype 
Teeth were not illustrated by Sues and Baird for the holotype 83.1 specimen, who reported the mandible was articulated. The authors described two premaxillary, six maxillary and eight dentary teeth. That low number of teeth point toward a procolophonid ancestry. The upper anterior four teeth are described as incisiform with bluntly conical crowns that are rounded in cross section. The first premaxillary tooth is reported to be much larger than the other teeth. A large medial pmx tooth also points toward a procolophonid ancestry, as we’ve already seen with Colobomycter. In Elginia (Fig. 3)  the many small teeth are slightly constricted at the base and serrated at the crown as in other pareiasaurs.

Figure 4. Elginia colorized in four views. Note the rotation of the tabulars to the dorsal skull.

Figure 4. Elginia colorized in four views. Note the rotation of the tabulars to the dorsal skull. Click to enlarge. Note the many similarities to the pareiasaur-like restoration of Scoloparia. 

Nuchal osteoderms
Sues and Baird noted “nuchal (neck) osteoderms” preserved posterior to the skull in the 83.1 holotype of Scoloparia. Cisneros (2008) reports osteoderms have only been found in Sclerosaurus and Scoloparia. Since Sclerosaurus nests here as a pareiasaur, that means no other procolophonids have osteoderms. Hmmm.

Reversals in the skull roof of pareiasaurs 
In the large reptile tree pareiasaurus are sisters to turtles (all derived from Stephanospondhylus) and bolosaurids, all derived from Milleretta. In Stephanospondylus (Fig 5) a reversal takes place in which the postparietals (or are they tabulars?) rotate to the dorsal surface of the skull and the supratemporals develop small horns. These traits usually appear on pre-amniotes.

Figure 2. Stephanospondylus skull in two views. Note the rotation of the post parietals to the dorsal skull along with the transformation of the supratemporals into small horns.

Figure 5. Stephanospondylus skull in two views. Note the rotation of the post parietals to the dorsal skull along with the transformation of the supratemporals into small horns.

This dorsalization of the tabulars
becomes even more apparent in pareiasaurs (Fig. 2) and Elginia (Fig. 4). If the purported nuchals of Scoloparia are actually large supratemporals, tabulars, and opisthotics, then it’s a pareiasaur. If so, a foramen magnum is also present topped by a supraoccipital and two flanking exoccipitals. What a quandary!

Not quite enough to go on
I am working from a 2D line drawing here (from Sues and Baird 1998), not a photograph. So I await images of the teeth and any other data that may come down the pike. If new data ever comes in, I will let you know. For now, can’t tell if we’re dealing with autapomorphic nuchal osteoderms on a procolophonid or dorsalized tabulars and an occiput on a pareiasaur.

References
Cisneros JC 2008. Phylogenetic relationships of procolophonid parareptiles with remarks on their geological record. Journal of Systematic Palaeontology): 345–366.
Sues HD and Baird D 1998. Procolophonidae (Reptilia: Parareptilia) from the Upper Triassic Wolfville Formation of Nova Scotia, Canada. Journal of Vertebrate Paleontology 18:525-532.