Not Sure If a Chaoyangopterid … or Just a Basal Azhdarchid

So, there’s a little something interesting that popped up while doing research on the systematics of pterosaurs. First, There’s not a whole lot we know about the skulls of a particular group of pterosaurs, the Chaoyangopteridae; and second, there may be more than a little uncertainty about the relationships OF the Chaoyangopteridae. Before I continue, let me introduce my protagonist:

A generalized skeleton in lateral view of a chaoyangopterid, in the Habib quad launch position, body horizonal. Skull is mostly illustrated from Shenzhouopterus chaoyangensis (Lü et al., 2008).

A generalized skeleton in lateral view of a chaoyangopterid, in the Habib quad launch position, body horizonal. Skull is mostly illustrated from Shenzhouopterus chaoyangensis (Lü et al., 2008).

(Quick note: Scott Hartman has begun illustrating his pterosaurs in this posture, with perhaps a bit of an upward tilt; I initially disagreed with this posture for skeleton presentation and preferred a full quadrupedal walk posture, but I’ve come to accept that this “sexier” posture is more generally useful for illustrating these animals both dynamically and in their dual habitat — ground and air — and additionally conveys the novel and fundamentally sound idea of the quad launch for getting off the ground.)

Chaoyangopteridae is the name given to a small (but seemingly growing) group of pterosaurs deeply nested into the Pterodactyloidea and which have been recovered almost exclusively as the sister taxon on Azhdarchidae. There are currently between 4-6 taxa included in this group, principally diagnosed by their unusually large and caudally-extending nasoantorbital fenestrae, and by a remarkably slender, shallow premaxilla above the fenestra, which apparently provided no dorsal cranial crest above the snout: Chaoyangopterus zhangi, Shenzhoupterus chaoyangensis, Jidapterus edentus and Lacusovagus magnificens, but may also include Eoazhdarcho liaoxiensis and Eopteranodon lii. Chaoyangopterids are apparently large-headed, long-legged, and generally devoid of a cranial crest. We’ll get to that last bit later (I can see all your eyes staring up at the reconstruction!). When it comes to skeletal preservation, chaoyangopterid bodies tend to preserve well the forelimbs and shoulder, much of the dorsal and cervical vertebrae, and most of the hindlimbs, but not much else. Thus it is somewhat speculative to reconstruct them of these parts, and so the greyed regions in the illustration are all that … speculation. But it is the relationship to azhdarchids that we turn, for some understanding of their evolution.

We all know what azhdarchids are, of course, with special reference to the only good “decent” skeletons being from the Javelina Formation of Texas, recovered in the 1970s from Big Bend National Park and were referred to as “Quetzalcoatlus sp.” by Kellner & Langston (1996) when the described several skeletons, including beautifully-preserved skulls. But the first azhdarchids found were from Jordan, in the Middle East, and were named Titanopteryx philadelphiae by Arambourg in 1959, based on some ridiculously elongated cervical vertebrae; a fluke in nomenclature — that is, there was an insect named out there already called Titanopteryx — forced renaming the material (by Nessov and associates, in 1987) to Arambourgiana. Lev Nessov in 1984 named a new group of pterosaurs with ridiculously long cervical vertebrae and edentulous beaks as Azhdarchidae, founded then on a series of pterosaur fossils scattered in a bonebed that seemed, at the time, to belong to a singular type of animal; thus, was born Azhdarcho lancicollis (the “spear-necked dragon,” after a Persian mythical spirit resembling a serpent, Dahaka). Since then, most azhdarchids have been described on the basis of one major defining feature: the extreme elongation of the mid-cervical vertebrae. This has allowed other remains, such as the partial skull of Hatzegopterus thambema (Buffetaut et al., 2002) to supplement what we began to think of as “azhdarchids.” A substantial number of new taxa have been named in recent years to Azhdarchidae, predicated on either a shallow, tapering rostrum indicating great size and/or the extremely long necks of these pterosaurs, and from all continents save South America, Antarctica and Australia. They are almost exclusively Late Cretaceous in age, though they span the range, with most concentrated in the Cenomanian through Maastrichtian.

By contrast, chaoyangopterids are known almost exclusively from the Barremian through to the Albian, to the Early Cretaceous, in which they seem to directly precede the azhdarchid radiation. Further, chaoyangopterids, when preserved well-enough, possess relatively elongated cervical vertebrae, but none of them so long as to even remotely compare to those of azhdarchids in which they are known. So, chaoyangopterids have “medium” length necks. This is comparable to other members of the Pterodactyloidea exclusive of Ctenochasmatoidea and Ornithocheiroidea (which Kellner termed Tapejaroidea, but Unwin termed [excluding only Ornithocheiroidea] Lophocratia). Yeah, it’s a big mess. Direct and broad application of phylogenetic nomenclature and definitions — and their acceptance — will go a long way to resolving these things; and it would be useful NOT to use taxonomy based on the Linnaean System, which favors use of “family”-grade stems in most names.

Dsungaripteridae and Tapejaridae (which I will casually associate with Thalassodromidae) for decent measure have associated with them skeletons which including virtually the entire cervical series, in Dsungaripterus weii and an undescribed skeleton which forms the basis of most skeletal reconstructions of Tapejara wellnhoferi you will find anywhere. In these, the necks are particularly short, and not as long as in Ctenochasmatoidea, but more comparable to Ornithocheiroidea, including Pteranodon longiceps. Thus, it would be the plesiomorphic condition for these pterosaurs, and that elongation (in Ctenochasmatoidea and the chaoyangopterid-azhdarchid clade) were secondary and convergent. I posit from this observation that chaoyangopterids represent a phase of elongation, and it is here where the title of this post comes in: That chaoyangopterids represent the basal stock from which azhdarchids arise, and that the Chaoyangopteridae (as generally conceived) is paraphyletic with regards to Azhdarchidae.

If we care a little about keeping the name “Azhdarchidae” neat and narrow, and restricted to expression of certain features of the skull or neck, we may also say that Chaoyangopteridae represents the basal stock from which other members of Tapejaroidea (least-inclusive clade containing Tapejara wellnhoferi, Dsungaripterus weii, and Azhdarcho lancicollis), or Azhdarchoidea (least-inclusive clade containing Azhdarcho lancicollis and Tapejara wellnhoferi), or Neoazhdarchia (most-inclusive clade containing Azhdarcho lancicollis, but not Tapejara wellnhoferi); Unwin’s Lophocratia was defined as the least-inclusive clade containing Pterodaustro guinazui and Quetzalcoatlus northropi, but I would here suggest that that second taxon be Azhdarcho lancicollis [under the premise that nomenclature founded on a “genus” such as Azhdarchidae use its type species in any coordinated definitions], but it should be further noted that in more recent phylogenetic analyses this clade is often synonymous with Pterodactyloidea, even though it was erected to distinguish taxa in which a mid-length cranial crest is present.

Postcrania of chaoyangopterids and azhdarchids may be virtually impossible to tell apart, aside from the cervical vertebrae, but what about the skulls? All chaoyangopterids are known from at least SOME skulls, and all of them preserve enough of the rostrum that, when describing Shenzhoupterus chaoyangensis, Lü Junchang and colleagues named the Chaoyangopteridae to emphasize these features. First, they noted the immensely large nasoantorbital fenestra, so large it seemed that the fenestra would extend caudally past the level of the jaw joint when the skull is horizontal; but also in the very slenderness of the supra-nasoantorbital fenestra bar, formed here — presumably — completely from the premaxilla. Jidapterus edentus soon followed, then Lacusovagus magnificens, all of them with slender, shallow rostra and a seemingly humongous nasoantorbital fenestra. Key to the extent of the fenestra is offered only in Shenzhoupterus chaoyangensis: the posterior region of the skull is either obscured, crushed, or missing in all other taxa. But certain features of the skull in Shenzhoupterus chaoyangensis suggest that the skull around the posterior end of the nasoantorbital fenestra is incomplete, and indeed the lacrimal and nasal portions of the dorsocaudal margin of the fenestra are missing, suggesting that the extent of the fenestra is artificial. This is, of course, provisional, and the skull is so far crushed in the holotype (HGM 41HIII-305A) as to make clear distinctions impossible. Cranially, chaoyangopterids can be distinguished from azhdarchids by the shape of the superior bar of the nasoantorbital fenestra, which in the former is evenly shallow along its length, while in the latter it is deeper rostrally and diminishes caudally. The former condition seems also present in Thalassodromidae, while the latter condition is present in Tapejaridae. One can, with little imagination, guess why these taxa form a complex of interchangable sister-taxon relationships.

The Javelina material called Quetzalcoatlus sp. further compounds the issue of distinction between the two clades by presenting a shallow, thin rostrum rather than the deep, “heavy” one of other azhdarchids, and this raises the question of a clinal increase of rostrum depth toward one or two (or more) subgroups, the development of the superior bar tapering in more than one clade, and elongation of the cervical vertebrae (which we already know to be convergent and potentially linked to ecological exploitation: Higher reach, increased surveillance height, etc.). Similarly, ecological concerns may promote the features seen commonly between the clades, and thus that the rostrum shape is selected for, rather than the deeper, more thalassodromid-like one.

Systematics of pterosaurs are strengthening yearly, but it is a slow process. In 2002-2003, Unwin and Kellner both offered up character matrices of close to 100 characters, and only recently with Andres, Ji and Lü in further analyses has this number effectively topped 110; Brian Andres’s thesis includes 182 characters, and it was produced in 2010! At this rate, we may barely top 200, while dinosaur systematics, with no less complex evolution being represented, rushes forth in the multiples of hundreds. The quality of character choice, the sampling not merely of taxa but of characters, the correct identification of continuous characters and how to treat them in a matrix, but also how to interpret the results (not as a Grail of some revealed truth but rather a Well from which to gain perspective).

I suspect, but cannot substantiate at this time, that chaoyangopterids in general represent a plesiomorphic stock of pterosaurs resembling the more basal thalassodromids, and from which azhdarchids arise. The use of the name “Chaoyangopteridae” may be fundamentally confusing. Sampling of new taxa or of specimens from this group may reveal more complex history than so far shown, while correct coding of possibly erroneously interpreted anatomy (including down to that dreaded unknown — “?”) can produce diverging results. That said, it should be interesting to test this hypothesis, and consider that even when we have a decent amount of suggestions for a holophyletic arrangement, this may merely be the product of too-little information, much of to do with a lack of perspective.

Andres B. & Ji Q. 2008. A new pterosaur from the Liaoning Province of China, the phylogeny of Pterodactyloidea, and convergence in their cervical vertebrae.Palaeontology 51:453–469.
Dong Z.-m., Sun Y.-w. & Wu S.-y. 2003. On a new pterosaur from the Lower Cretaceous of Chaoyang Basin, Western Liaoning, China. Global Geology 22:1–7.
Kellner, A. W. A. 2003. Pterosaur phylogeny and comments on the evolutionary history of the group. in Buffetaut & Mazin (eds) Evolution and Palaeobiology of Pterosaurs. Geological Society of London, Special Publication 217:105–137.
Kellner, A. W. A. & Langston, W., Jr. 1996. Cranial remains of Quetzalcoatlus (Pterosauria, Azhdarchidae) from Late Cretaceous sediments of Big Bend National Park, Texas. Journal of Vertebrate Paleontology 16(2):222-231.
Lü J.-c. & Ji Q. 2005. New azhdarchid pterosaur from the Early Cretaceous of Western Liaoning. Acta Geologica Sinica 79:301–307.
Lü J.-c & Zhang B.-k. 2005. New pterodactyloid pterosaur from the Yixian Formation of Western Liaoning. Geological Revisions 51:458–462.
Lü J.-c., Unwin, D. M., Jin X.-s., Liu Y.-q. and Ji Q. 2010. Evidence for modular evolution in a long-tailed pterosaur with a pterodactyloid skull. Proceedings of the Royal Society, B: Biological Sciences 277:383-389.
Nessov, L. A. 1984. Pterosaurs and birds of the Late Cretaceous of Central Asia. Paläontologische Zeitschrift 1:47–57.
Unwin, D. M. 2003. On the phylogeny and evolutionary history of pterosaurs. in Buffetaut & Mazin (eds) Evolution and Palaeobiology of Pterosaurs. Geological Society of London, Special Publication 217:139-190.
Wang X.-l. & Zhou Z.-h. 2002. A new pterosaur (Pterodactyloidea, Tapejaridae) from the Early Cretaceous Jiufotang Formation of Western Liaoning, China and its implications for biostratigraphy. Chinese Science Bulletin 47:15–21.
Witton, M. P. 2008. A new azhdarchoid pterosaur from the Crato Formation (Lower Cretaceous, Aptian?) of Brazil. Palaeontology 51(6):1289-1300.

This entry was posted in Biological Comparison, Biology, Paleobiology and tagged . Bookmark the permalink.

8 Responses to Not Sure If a Chaoyangopterid … or Just a Basal Azhdarchid

  1. Very interesting hypothesis and wonderful post. Just a question: is that skeletal diagram of yours a general chaoyangopterid or a more secific genus?

    “A substantial number of new taxa have been named in recent years to Azhdarchidae, predicated on either a shallow, tapering rostrum indicating great size and/or the extremely long necks of these pterosaurs, and from all continents save South America, Antarctica and Australia.”
    Not exactly true. Aerotitan, recently described from a piece of rostrum, is an azhdarchid recovered from South America.

  2. Mark Witton says:

    Hi Jaime,

    There are a number of differences between chaoyangopterids and azhdarchids. So far as I can see, these taxa do not share any features apart from general azhdarchoid plesiomorphies, whereas thalassodromids and azhdarchids share a number of derived postcranial characteristics including the possession of several derived wing proportions, and distinct humeral anatomy (I have a paper on this in the works). As such, at least some azhdarchids and chaoyangopterids postcranial components, and perhaps many, can be readily distinguished. I suspect azhdarchids and chaoyangopterids are in the same clade (along with Thalassodromidae), but I do not see evidence for direct evolution between them.

    Also, although the Shenzhoupterus holotype is a mess (and the images in the paper aren’t great), I can vouch for the interpretation by Lu et al. after spending a couple of hours looking at the specimen in 2010. I don’t think it’s missing as many bits as you suggest here (I am very sceptical about the your ‘broken’ crest margin, in particular), the back of the skull is probably higher, and the orbital region more akin to those of thalassodromids and azhdarchids than you have depicted.

    • Mark, thanks for the reply.

      I appreciate the response. Sadly, I don’t have extensive material on the ridiculous skulls of these critters, and was trying to generalize them. Part of the reconstruction stems from a hesitancy to allow taphonomy to tell me that the current preservation reflects the original state; I am reminded specifically of Tapejara wellnhoferi and broad acceptance of the diagnostic value of the somehow “floating” premaxilla above the remainer of the braincase because in various specimens of species of tapejarid this bar is loose. I figured it was safer to propose a more “moderate” cranial reconstruction, one in which perhaps Jidapterus edentus</i? might agree with, where the NAOF does not extends further caudally than the rostral margin of the orbit. That said, this is really a quick skeletal illustration, and is meant to convey the idea of a chaoyangopterid, it is not perfect though the proportions are accurate and thus does not convey the correct shapes etc. of any specific species. As with Thalassodromeus sethi, the caudal section of the skull appears to display an incomplete crest, where in sethi this is a series of V-shape notches cut out of the crest (I’ve reconstructed the skull as essentially rounded on this margin, as in the restoration here). I’ve not spent nearly as much time with these as you have, so I freely admit my ignorance, but in this I consider the implication of taphonomic loss of material to be more parsimonipous than the idea of natural notches. So, too, the back of the Shenzhoupterus chaoyangensis skull, which has a broadly concave caudal margin as though a circular “cut” were taken out, with seemingly either two caudal spurs of the occipital “plate” (supraoccipital AND premaxilla/parietal, but not both as a single crest as in all other pterosaurs?). I will of course concede your points of postcranial proportions; I admit for the moment that my observations on this were based on the inability of recent cladistic analyses to agree as to their placement, getting either thalassodromids or chaoyangopterids at the base. Nonetheless, it would also not matter were either of them the sister taxon to azhdarchids, as the same problems of convergent limb and neck porportions become involved either way; I merely think it more interesting to note the relatively longer necks in chaoyangopterids compared to that one thalssodromid skeleton that exists, and azhdarchids. This perspective comes from having immersed myself in the guts of several of these to ascertain the relationships of a particular specimen.

  3. Key overlooked difference between azhdarchids + chaoyangopterids on the one hand and tapejarids on the other is that the former have parallel rostral edges as seen ventrally or dorsally, like two yardsticks coming together because they were derived from dorygnathids. Tapejarids are closer to shenzhoupterids, dsungaripterids, eopteranodontids and germanodactylids in having a sharp tipped rostrum. Don’t forget the short neck verts with high neural spines on the tapejarids and the large square pectoral complexes, rather than the tiny dorygnathid complexes. Those and a host of other traits, including pedal and manus proportions derived from tiny precursors that never seem to be ‘worthy’ of current phylogenetic analyses appear to stand in the way of current understanding. Details can be seen at the appropriate pages on (like, and their relationships are shown at the completely resolved pterosaur tree found here:

    With regards to the forelimb leap pictured above, note that the triceps (from shoulder to elbow), are only half the length and therefore one eighth the volume of the thigh muscles driving the hind limbs. Not to mention that that the wings in your reconstruction need to unfold (which will take a lot of space and a fair amount of time) and rise at least above the shoulder prior to the first downstroke and imminent crash. A simple bird-like launch solves all such problems.

    Take it from me, and I should know, radical thinking will only come back to bite you.

    Best regards, Dave

    • I will not debate the phylogenetics on this right now, Dave. I will say though that I have at least two papers in preparation or submission that deal with the phylogenetics in this or related areas, and that the structure of the snout is of particular focus in BOTH of these, thus that there is some emphasis on clarifying the rostrum anatomy. I will say that the shape of the snout in lateral and ventral views is NOT cut and dry, you don’t just get one clade with narrowly-approaching jaw tip, versus broadly approaching, etc.; in all Tapejaroidea/Azdarchoidea/Neoazhdarchia, the lateral margins of the snout approach the tip at a very low angle, even excepting the unusually short snout of Tapejara wellnhoferi. This is even true for Dsungaripteridae, Pteranodontidae and Nyctosauridae; it may not be true of Quetzalcoatlus sp., but that’s not an issue — it’s one internal outlier, nested among animals with very-narrowly-approaching snouts. The functional anatomy of a slender or shallow snout should also give warning as to the phylogenetic quality of characters of the snout in shape, and thus that the characteristics of them (especially when considering that some of your “dorygnathids” have squished skulls) are not as useful as you might think.

  4. forex says:

    Fuck you all Forex bustards

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