Friday, 12 September 2014

Jidapterus: the pterosaur they got upside-down!

The first report of the pterosaur Jidapterus was by Dong, Sun & Wu 2003. Jidapterus is a Chinese pterosaur from the Jiufotang Formation of western Liaoning, making it Aptian in age. It belongs to the taxonomic family Chaoyangopteridae, which is itself in Azhdarchoidea. As an azhdarchoid it has a typically short last wing phalanx (knuckle); a long wing metacarpal (wrist bone) relative to the rest of the wing; a long neck relative to the body; and a large skull, lacking teeth. Initially Jidapterus was not assigned to a family, as Chaoyangopterus had only just been described itself, and it was considered to be a nyctosaur (Wang & Zhou 2003). It wasn’t until Shenzhoupterus Lü, Unwin, Xu & Zhang 2008 was described that the family Chaoyangopteridae was erected. Lü et al. (2008) defined Chaoyangopteridae as being azhdarchoids with an extremely thin premaxilla bounding the nasoantorbital fenestra (the external nares) [NAOF], while the NAOF itself extended back to the jaw joint – among other anatomical features. While the posterior portion of the skull in Jidapterus is poorly preserved, the specimen’s very large/expanded NAOF makes it difficult to argue against its placement in Chaoyangopteridae.
Jidapterus was reconstructed (Dong et al. 2003) as having a skull with a beautifully straight dorsal margin, whilst having a curved occlusal surface (biting surface). However, this is very different to Chaoyangopterus, Shenzhoupterus and Lacusovagus, which represent all of the good skull material within Chaoyangopteridae (see image below).
Reconstruction of Jidapterus by Dong et al. 2003

Skulls of chaoyangopterids A) Chaoyangopterus; B) Shenzhoupterus; C) Lacusovagus

Unfortunately I am limited by the images available to me and the content of the original paper (i.e. Dong et al. 2003) that I am able to translate, which is why I am now making the disclaimer that this is not incontrovertible fact, but some ideas that suggest the material needs to be revisited. From what I can tell looking at the images in the paper and the information that new fossil discoveries have provided me with, Jidapterus is upside-down!
Original figure of Dong et al. 2003: Note the label for maxilla (M) which is only present on the occlusal surface of the pterosaur rostrum.

My reasoning behind it - other than it being inconsistent with other chaoyangopterids - is as follows.
            1.       The orbit in azhdarchoids is lower than the dorsal margin of the NAOF.
            2.       The curved skull surface of the skull is longer than the straight surface, which would make the lower jaw significantly  shorter than the upper if the original reconstruction is correct.
            3.       It appears that part of the jugal (the bone defining the ventral margin of the orbit) is intact on the end of the “dorsal spar of the premaxilla”, which is likely to be the maxilla.
            4.       The neck is in articulation and is directed towards the “dorsal surface of the skull”; you would expect it to be directed towards the ventral surface.
So, bearing all that in mind the skull is most likely reconstructed upside-down, and in fact reconstructed very wrongly. I’m pretty sure that a more detailed examination of the material would reveal this and solidify our understanding of chaoyangopterids. See my new reconstruction below.
Reconstruction of Jidapterus
Morphologically, it appears more similar to Chaoyangopterus and Lacusovagus than Shezhoupterus.

Note: It seems that Lü et al. (2008) and Witton (2008), like myself, considered Jidapterus to have a “straight” occlusal surface to its rostrum, but provide no discussion and only give the reference to the original paper which clearly figures the straight margin as the dorsal process of the premaxilla.

Below is the original photo of the Jidapterus specimen with a scale bar, inferred from the wing phalanx lengths provided by Dong et al. (2003).

Dong, Sun & Wu. 2003. "On a new pterosaur from the Lower Cretaceous of Chaoyang Basin, Western Liaoning, China". Global Geology 22:1-7
Lü, Unwin, Xu & Zhang. 2008. A new azhdarchoid pterosaur from the Lower Cretaceous of China and its implications for pterosaur phylogeny and evolution. Naturwissenschaften. DOI 10.1007/s00114-008-0397-5
Wang & Zhou. 2003. "Two new pterodactyloid pterosaurs from the Early Cretaceous Jiufotang Formation of Western Liaoning, China". Vertebrata PalAsiatica 41 (1): 34–41.
Witton. 2008. A new azhdarchoid pterosaur from the Crato Formation (Lower Cretaceous, Aptian?) of Brazil. Palaeontology 51:1289-1300.

Monday, 7 July 2014

Flugsaurier 2015

I am proud to be a member of the Flugsaurier 2015 committee. Flugsaurier 2015 is the 5th international symposium on pterosaur research. The conference is being hosted by the University of Portsmouth's School of Earth and Environmental Sciences (SEES), in the historic coastal city of Portsmouth, UK.

SEES is where I work on my pterosaur research along with the other members of the University of Portsmouth's Palaeobiology Research Group. Over the years my colleagues and supervisors have collaborated with other researchers to describe and name several new pterosaurs from the UK, Morocco and Brazil. We are continuing this tradition of pterosaur research and some exciting new projects are underway and promise to make worthwhile features of the conference.

Past Flugsaurier meetings have been well attended by an international delegation. Revolutionary theories have been put forward to the delegation during these meetings and excellent new specimens from all corners of the globe have been unveiled. We expect Flugsaurier 2015 to continue this trend.

The conference will offer two days of talks and poster presentations on cutting edge pterosaur research, workshops on issues effecting contemporary pterosaur research and opportunities to examine pterosaur specimens.We will be running two pterosaur-themed field trips: a pre-conference trip to the Isle of Wight on August 25th, and a post-conference weekend field trip to the Jurassic Coast World Heritage Site on August 29th and 30th. The second field trip will be run jointly with SVPCA (Symposium of Vertebrate Palaeontology and Comparative Anatomy) 2015, which is being held immediately after Flugsaurier in the neighbouring city of Southampton, August 31st to September 4th.

To find out more about the conference and to keep up to date, navigate to the following links:
Sign-up to the mailing list

Tuesday, 11 March 2014

A new enigmatic fossil crocodyliforme from the Isle of Wight, UK: Koumpiodontosuchus aprosdokiti Sweetman, Pedreira-Segade and Vidovic 2014

Mesozoic Monsters: a whole heap of vertebrate palaeontology, straight from the desk of researchers in the field. This is the title and tagline of my blog. I had every intention – when I started this blog – of regularly updating my readership with little titbits of my research, but I soon learned that it is not that simple and ideas need protecting and publishing first. So this is to be the first article that is submitted in the true spirit of MM.

The newly described Koumpiodontosuchus aprosdokiti Sweetman, Pedreira-Segade and Vidovic 2014 is a small crocodyliforme, neosuchian, bernissartiid from the Isle of Wight, UK. The specimen was found a couple years ago on the foreshore of Yaverland beach, Isle of Wight. The associated sediment suggested that the material had come out of the Wessex Formation, Wealden Group. The specimen, which is the holotype (type material) and only specimen of Koumpiodontosuchus is represented by a skull which was broken in two and found by separate individuals, but they were married together by some “serendipity” (SCW in Sweetman et al. 2014). I had the good fortune of working on this specimen, and I was tasked with placing it in the crocodile evolutionary tree, which I will discuss further below.
A palaeontological reconstruction of the Wessex Fm. Biota, by Mark Witton, reproduced with his permission.
Koumpiodontosuchus is a small crocodyliforme at an estimated 66cm in total length – the skull is 11.2cm in length. The dentition is distinctive and alerted us to the natural affinities of this small croc. It has a tribodont dentition in the distal portion of the jaw (back of mouth). Tribodont teeth are bulbous, and kind of onion-shaped. Clearly these teeth are involved more in digestion rather than prey capture. The mesial teeth (front) are somewhat procumbent (forward pointing) simple cones, followed by some pseudo-canine teeth, one of which projected up from the lower jaw and sat in a notch in the side of the upper jaw, just behind the nares (nostrils). This morphology is typical of a previously monotypic family (only contained one genus), the Bernissartiidae Dollo, 1883. Thus, it would be expected that a cladistic analysis would have no problem placing Koumpiodontosuchus with the type genus of the family Bernissartia fagesii “Known from contemporaneous strata” (Sweetman et al. 2014) in Belgium and Spain. However, there are some distinguishing features of Koumpiodontoshuchus which are not only distinct from Bernissartia, but also convergent upon Eusuchia. Namely, the position of a perforation named the choana within the pterygoid bones, which is one of the defining characteristics of Eusuchia.
This image is made available by APP through the creative commons licence. Originally produced by UP-S & SCW in Sweetman et al. 2014.

The specimen was first made public at SVPCA Oxford, but at that time it became clear – in light of the above convergences – that there were some unusual features this skull possessed which required further exploration with cladistic methods, and so I was signed up – hence the one and a half year hiatus.
The presence of compound characters in cladistic analyses means that a systematist (someone who performs these analyses) may not be telling the computer what they believe they are (Brazeau, 2011). As such character conflict may be artificially high, causing some natural groups to lack support, while other tentative groups gain support. Due to this and the chimaera-like situation that Koumpiodontosuchus found itself in meant that a whole new ‘Brazeau-acceptable’ analysis had to be constructed, picking and choosing, or altering characters from published analyses. So, in order to assess the systematic position of Koumpiodontosuchus an analysis coding 36 taxa for 242 characters was run in TNT, using a NT search. The result – after running more procedures to find the maximum number of MPTs (most parsimonious [best fitting] trees) – was a strict consensus of 17 trees with a length of 862 steps. A single tree was generated using implied weighting, which is the tree we finally presented. The logic behind this is that implied weighting changes the ‘impact’ of conflicting characters whilst the analysis runs, thus diminishing the effects of homoplasy (convergence of characters due to reasons other than a natural affinity): perfect for this scenario. The result (see cladogram) was that the Bernissartiidae was found as the sister-group to Rugosuchus, goniopholidids and Eusuchia. Interestingly, Thalattosuchia – a major group of marine crocodiles (see my article on Dakosaurus) – was found within Tethysuchia, a grouping that previously contained dyrosaurs and pholidosaurs, exclusive of thalattosuchians. Worryingly, the presence of the choana within the pterygoids of Koumpiodontosuchus, and its placement outside of Eusuchia leaves the clade of ‘true crocodiles’ wanting a definition. Salisbury et al. (2006) were happy to define the clade based on Isisfordia being the most basal member of the family, but according to this analysis that would exclude a lot of traditional Eusuchians. Brochu (1999) thought that Eusuchia could be defined by the most recent common ancestor of Hylaeochampsa and Crocodylia, but this is also an unreliable solution due to the shifting position of Hylaeochampsa between many analyses.

This image is made available by APP through the creative commons licence. Originally produced by SUV in Sweetman et al. 2014.
So what do we do? Abandon Eusuchia? Pfsh... I’m sure I don’t know. Not right now anyhow. The analysis I performed was mainly based upon cranial characters, with a very small subset of specimens. So a more comprehensive analysis in the same vein (really scrutinising characters for what they will tell the computer, in the light of Brazeau’s [2011] work) is required. I am currently in two minds as to whether I should pursue this point of enquiry or whether I should leave it to more experienced/more interested individuals. Who knows, it could be one of my next big posts.