Back to Cuspicephalus: Germanodactylid? or Wukongopterid?

Earlier we looked at the gracile skull of Cuspicephalus which nested with Germanodactylus B St 1892 IV 1 in the large pterosaur tree.

Today we revisit this taxon after the publication of Witton et al. 2015, which attempted to related Cuspicephalus to Darwinopterus and the wukongopterids.

Cuspicephalus scarfi

Figure 1. Cuspicephalus scarfi. Click to enlarge. Note the round exoccipital process at the back of the skull. Germanodcatylids have these. Wukongopterids do not. Witton thinks that bone is an artifact.

I have rarely seen a paper with such a bogus foundation…

  1. Witton et al. support the ‘modular’ evolution of pterosaurs at the base of the Pterodactyloidea. Earlier we learned that with the simple addition of taxa (which other workers continue to avoid) there are four origins for pterodactyloid-grade pterosaurs, all following phylogenetic miniaturization, a process that happens often in reptile evolution. We also learned that there is no such thing as ‘modular’ evolution with half the body evolving and waiting for the other half to catch up. In the case of wukongopterids, the other half never developed pterodactyloid-grade traits. Wukongopterids were a terminal taxon. The non-modular evolution of long tails to short tails happened several other times at several other nodes in the pterosaur cladogram (including in the anurognathids).
  2. With several distinct genera and specimens now nesting close to Darwinopterus robustus within the Wukongopteridae, no attempt was made to figure out which of these specimens were more basal and which were more derived — as shown in the large pterosaur tree.
  3. Witton et al. support a monophyletic “Monosfenestrata” which, to them, includes pterodactyloids + wukongopterids, a tree topology that is not supported by any other published studies. In the large pterosaur tree, wukongopterids, like anurognathids, convergently developed some pterodactyloid-grade traits and not others, then left no descendants.
  4. Witton et al. did not produce their own skull tracings, but rely on cartoonish and inaccurate versions of prior work by others (apparently often by Bennett 1996). Few to no skull sutures are shown and certain inaccuracies are present.
  5. Witton et al. are not critical of the cladistic work of others (Andres et al., 2014; (Lü et al., 2010; Tischlinger and Frey, 2014), nor do they offer support for the matrix they preferred (Unwin 2003). Seems less scientific than one would like to see here. Or did they not want to do the work? Or make enemies?
Figure 2. Cuspicephalus compared to Darwingopterus and to Germanodactylus, all to scale.

Figure 2. Cuspicephalus compared to Darwingopterus and to Germanodactylus, all to scale. Click to enlarge. The large reptile tree nests Cuspicephalus with Germanodactylus. Witton et al. report a closer relationship to Darwinopterus. The presence of large exoccipital ‘ears’, an extended cranium, a pointed rostrum, a pointed ventral orbt, an alignment of the rostral crest and antorbital fenestra anterior margin all argue for the present hypothesis. The longer antoribital fenestra developed by convergence in Darwinopterus and Cuspicephalus.

wukongopterid skulls are indeed very similar to those of germanodactylids (Fig. 2). Both clades also offer a wide variety of shapes and sizes.

With regard to a key trait
in Cuspicephalus scarfi (MJML K1918) from Witton et al. 2015: The exoccipital processes are unexpanded: they look relatively large on MJML K1918, but this is largely an artefact of distortion around the occipital region, and they are not as prominent as those of Germanodactylus or dsungaripterids.

in Cuspicephalus the exoccipitals are just as big, if not bigger relative to skull height (Fig. 2).

In the large pterosaur tree
Cuspicephalaus nests with B St 1892 IV 1, n61 in the Wellnhofer (1970) catalog, which nests with two headless taxa, Wenupteryx (MOZ 3625) and the so-called “Crato azhdarchide (SMNK PAL 3830)”

From Witton et al. “Our assessment suggests that wukongopterid skulls can be distinguished from other Jurassic monofenestratans by not only lacking the well-documented cranial  synapomorphies of pterodactyloid clades, but also through a unique combination of characters (Darwinopterus, Gemanodactylus and Cuspicephalus = D, G and C):

  1. Striated bony crest lower than the underlying prenarial rostrum, with sloping anterior margin – actually lower in G.
  2. Anterior crest terminates in the posterior region of the prenarial rostrum, closer to the anterior border of the nasoantorbital fenestra than the jaw tip – note the crest starts more anteriorly in D
  3. Reclined, but not sub-horizontal, occipital regions leans more in G.
  4. Piriform (pear-shaped) orbitbut in G and C the orbit is sharply angled ventrally
  5. Convex anterodorsal orbital marginmore convex in G + C.
  6. Short nasal processonly in D.
  7. Unexpanded exoccipital processesonly in D.
  8. Concave dorsal skull surfacenot on G, D or C.
  9. Straight ventral skull surfacepresent on G, D and C.
  10. Nasoantorbital fenestra over 50% of jaw length – on D and C
  11. Small, equally sized alveoli – only on C, larger teeth on D and G.
  12. First alveolus pair located on anterior face of jaw, with mandible over-bitten by first premaxillary tooth pair – present on G, D and C
  13. Regular tooth spacing – only on D
  14. Interalveolar spacing generally greater than tooth length – only on D
  15. Dentition extends under anterior half of the nasoantorbital region – only on G and C
  16. Relatively slender, sharply pointed conical teeth – only on C.

Cuspicephalus is Kimmeridgian (Late Jurassic) in age. So is Germanodactylus (Kimmeridigian/Tithonian). Darwinopterus is late Middle Jurassic (Bathonian/Oxfordian) in age. No wukongopterids are found in Late Jurassic deposits. So far…

I can see why there is confusion here. 
The skulls are very similar in overall morphology. But the weight of evidence appears to lend weight to a Germanodactylus relationship for Cuspicephalus. If Witton et al. had made more precise tracings and reconstructions, if they had used a valid tree topology that included tiny pterosaurs, if they had not discounted the presence of exoccipital processes on Cuspicephalus, then I think they would have come up with a nesting that echoed that of the large pterosaur tree.

An outlandish suggestion based on a cladogram
We have a large germanodactylid skull without a body (Cuspicephalus) and we have a large germanodactylid post-crania without a skull (the Crato Azhdarchide). Although they are separated somewhat in time, they are sister taxa. Wonder how well the real skull and real post-crania would match up with these two…

Diopecephalus = P. longicollum = Ardeadactylus. Normannognathus is in the box in the lower left.

Figure 3. Witton et al. also attempted to resolve the relationships of Normannognathus without success. Here it is in the box at lower left. Phylogenetic analysis nests it with Diopecephalus = P. longicollum = Ardeadactylus.

Witton et al. also considered the problem of the placement of Normannognathus (Fig. 3). Earlier we looked at the phylogenetic relationships of Normannognathus (Buffetaut et al. 1998;  MGC L 59’583) known from a toothy, curved rostrum and crest. While Witton et al. considered the problem too difficult to solve, several years ago Normannognathus was matched to the big Pterodactylus longicollum (SMNS-56603, No. 58 of Wellnhofer 1970), which was not considered by Witton et al.

Andres B, Clark J, Xu X. 2014. The earliest pterodactyloid and the origin of the group. Current Biology 24: 1011-1016.
Bennett SC 1996. Year-classes of pterosaurs from the Solnhofen limestones of Germany: taxonomic and systematic implications. Journal of Vertebrate Paleontology 16:432–444.
Lü JC, Unwin DM, Jin X, Liu Y, Ji Q. 2010. Evidence for modular evolution in a long-tailed pterosaur with a pterodactyloid skull. Proceedings of the Royal Society B 277: 383-389.
Tischlinger H and Frey E 2014. Ein neuer Pterosaurier mit Mosaikmerkmalen basaler und pterodactyoider Pterosaurier aus dem  Ober-Kimmeridgium von Painen (Oberpfalz, Deutschland) [A new pterosaur with moasic characters of basal and pterodactyloid Pterosauria from the Upper Kimmeridgian of Painten (Upper Palatinate, Germany)]. Archaeopteryx 31: 1-13.
Witton MP, O’Sullivan M and Martill DM 2015. The relationships of Cuspicephalus scarfi Martill and Etches, 2013 and Normannognathus wellnhoferi Buffetaut et al., 1998 to other monofenestratan pterosaurs.



Leave a Reply

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

You are commenting using your 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.