The pterodactyl that fell down the back of the sofa, part 1: another ‘unexpected discovery’

It’s no secret that many palaeontological ‘discoveries’ aren’t made in the field, but are actually stumbled into by researchers working in museum collections. This is not surprising in the slightest: the flagship museums of many nations are rammed with - literally - millions of specimens. Some of these are virtually undocumented and unknown, even to experts in relevant fields, and require visiting researchers to be in the right frame of mind, to recognise and appreciate as something worthy of putting on record. It’s also well known that the preparation of many discoveries can take such a long time that the treasures brought back from exhibitions to exciting localities can sit, unknown, decades after decade. It is not inconceivable to imagine a whole career being forged by picking your way through the archives of big museums, looking at fossils that have been missed, unopened or in need of reappraisal.

Last week, I played this very game myself, by bringing attention to an overlooked pterosaur specimen held in the bowels of the Natural History Museum, London (Witton 2012 – free to download from PLoS ONE. Image, above, shows the title slide from a talk I've given on this research). The twist to my own version of this tale is that the element I described does not only belong to a very familiar pterosaur species, but a very familiar specimen. The short version of this story is that reappraising a long-forgotten component of a well-known British pterosaur suggests that its 100-year standard skull reconstruction is incorrect, which has obvious knock-on effects for its taxonomy and functional morphology. Those of you with lives to lead may as well log of here, but, if you have a lot of time to kill, read on over this series of posts for more back story and details.

The who

The animal in question is Istiodactylus latidens, a 4.2 m span ornithocheiroid from the Lower Cretaceous Vectis Formation, of the Isle of Wight (image, above, shows I. latidens launching, from my upcoming book. Some people may be happy to hear that the first draft has been submitted!). I. latidens is the largest species of Istiodactylidae and the only member of this group known to occur outside of Cretaceous deposits of China’s Liaoning region*. Istiodactylids are characterised by their muzzles of interlocking, razor-edged teeth and have been called the ‘duck-billed pterosaurs’ by some but, as we’ll see later, this analogy is plain daft: there is nothing at all duck-like about istiodactylid jaws). Istiodactylus has been known for a long time, though much of its early research history is murky. The documentation of its discovery, and early inventories of material referred to this species, are particularly vague. It’s not even known how pterosaur grandpapa Harry Seeley knew of I. latidens unusual teeth when he named the species in 1901, as the holotype specimen appears to lack jaw elements (‘latidens’ means ‘broad tooth’). Howse et al. (2001) suggested that an un-described skull in the University Museum of Zoology, Cambridge may belong to the holotype material, but this has not been confirmed. Seeley’s naming of the material was also of borderline validity, and some nomenclatural wrangling was needed to straighten out the taxonomy of the specimens we now know as Istiodactylus (Howse and Milner 1993; Howse et al. 2001).

*There have been several claims to the contrary, however. Bakker [1998] reported an alleged istiodactylid jaw from the Upper Jurassic Morrison Formation of Colorado, but this has not been accepted by the pterosaur community and seems to represent something more akin to Darwinopterus or a basal ctenochasmatoid. A reappraisal of that specimen is needed to say anything definite, though. Arbour and Currie (2010) named an Upper Cretaceous istiodactylid from British Columbia, Gwawinapterus beardi but, for reasons discussed in Witton (2012), I have considerable doubt that this material is pterosaurian, let alone an istiodactylid. There are reports, however, of a Cretaceous istiodactylid from Lebanon: I think a description is underway.

The what
I. latidens is amongst the best known istiodactylids of all and was, for much of the 20th century, one of the only pterosaurs represented by substantial, uncrushed three-dimensional material. Of the various specimens referred to I. latidens, one is preferentially discussed far more than the others: NHMUK R3877. Represented by an almost complete skull (see below) and a good portion of postcranial material, this specimen is the material that most associate with the name I. latidens. NHMUK R3877 was collected in 1904 from Atherfield Point on the Isle of Wight, and subsequently described and illustrated in detail by Reginald Walter Hooley (1913), a dedicated ‘amateur’ palaeontologist who collected and described many important specimens of Cretaceous reptiles from the Isle of Wight. Pterosaur workers have added little detail to the picture of I. latidens since Hooley’s day and, aside from a little nomenclatural juggling in the 90s and 00s, the picture of I. latidens has remained unchanged since the 1913 description. Hooley’s picture of NHMUK R3877 has become very familiar thanks to its continual discussion in pterosaurian technical literature, be it for anatomical comparisons or for use in functional morphology. We also still use Hooley’s original bone dimension estimates for NHMUK R3877, of which there are many: though well-preserved, few bones are complete. This latter issue is the reason for the waters around I. latidens and NHMUK R3877 finally being unsettled after a century of stillness. (Image above is Stafford Howse's life reconstruction of I. latidens, primarily based on NHMUK R3877. From Howse et al. 2001)

Ripples in the pond
Last June I travelled up to the NHM with neonate palaeontologists Kirsty Morgan and Georgia Maclean-Henry with a goal of photographing NHMUK R3877 for my book. The skull was my main priority, as the two skull pieces of NHMUK R3877 show details of istiodactylid skulls fantastically. One block shows the elongate, delicately-built posterior region, and the other contains both jaw tips, complete with smiling, interlocking teeth. These remains do not articulate, as the middle region has long been considered missing or, perhaps, only represented by useless scraps of bone. Generally, around 300 mm has been thought missing from the mid-jaw region, giving I. latidens a long skull length of 560 mm (Hooley 1913). On our trip, I ended up riffling through the many drawers containing NHMUK 3877 more thoroughly than usual and, in the odds-and-sods drawer, containing some of the less impressive bits and pieces of the specimen, I stumbled across this:

You’re looking at the c. 140 mm length of maxilla and portion of mandible from NHMUK R3877, a rather unimpressive collection of bones representing the mid-jaw length of the skull and lower jaw. There are only a few features worthy of mention, being the groove extending along the medial surface of the maxilla and the very shallow depth of the same bone, which measures 6 – 7 mm along its length. I must admit to having been rather ignorant of this third skull piece in the past, but Hooley knew of it, identified it and even figured it (Hooley 1913, Pl. XXXVII, Fig. 4). Until perhaps fairly recently**, however, it was in a rather unprepared state and of considerably less interest than the other, sexier bits of the fossil. Interestingly, I’m not alone in my ignorance: this portion of the skull that has never been incorporated into a reconstruction of I. latidens skull (despite at least four efforts that I know of) and has not been mentioned, to my knowledge, since Hooley’s brief description of it in 1913. Because I’m basically a child with an irrepressible urge to articulate broken fossils bones where possible, it didn’t take long for me to start wondering if this broken bit of jaw would fit with either of the other skull pieces. The answer was almost certainly yes, to both of them: it seems that a bridging element to the two NHMUK R3877 skull pieces was there all along, but had simply been forgotten or ignored.

**I’m unsure of the exact date of preparation, but I recently noticed that a photograph in Wellnhofer’s (1991) pterosaur encyclopaedia shows the specimen in the unprepared state. There is no date provided for the photo, but it does suggest that the specimen was left not prepped for several decades following Hooley’s description.

It must be said that I didn’t take this idea lightly: I not only asked my colleagues and NHM curator (and Pterosaur.Net contributor) Lorna Steel for corroboration of the fit, but sought further reassurance from David Martill and three pterosaur-studying PhD students before I believed my own eyes. After all, there is 100 years of intellectual inertia around the length of the NHMUK R3877’s jaw, and I figured that others would have found some flaw with their close association. Turns out that there wasn’t: for whatever reason, people had simply not put the material back together. The fit, it must be said, is not pin-point perfect, but the dimensions of the broken regions, the position of the maxillary medial grove, and the displacement of the dentary from the upper jaw are very close matches. This suggests that we’re actually only missing millimetres of the NHMUK R3877 jaw, not almost 300 mm. If this is the case, then the length of jaw between the two large skull pieces is only half that supposed for the preceding century, so we may seriously need to seriously overhaul our impression of the I. latidens skull. But should we be so hasty? After all, Hooley was no fool, so perhaps we need to consider his ideas in greater depth before rejecting his proposed jaw length. Perhaps there are other interpretations about the fit of this third element we could also consider. For that, you'll have to come back for part 2...

References

• Arbour, V. M. and Currie, P. J. 2010. An istiodactylid pterosaur from the Upper Cretaceous Nanaimo Group, Hornby Island, British Columbia, Canada. Canadian Journal of Earth Sciences, 48, 63-69.
• Bakker, R. T. 1998. Dinosaur mid-life crisis: the Jurassic-Cretaceous transition in Wyoming and Colorado. S. G. Lucas, J. I. Kirkland, & J. W. Estep. (eds.) Lower and Middle Cretaceous Terrestrial Ecosystems, New Mexico Museum of Natural History and Science Bulletin, 14, 67-77.
• Howse, S. C. B. and Milner, A. R. 1993. Ornithodesmus – a maniraptroan theropod dinosaur from the Lower Cretaceous of the Isle of Wight, England. Palaeontology, 36, 425-437.
• Hooley, R. W. 1913. On the skeleton of Ornithodesmus latidens; an Ornithosaur from the Wealden Shales of Atherfield (Isle of Wight). Quarterly Journal of the Geological Society, 96, 372-422.
• Howse, S. C. B., Milner, A. R. and Martill, D. M. 2001. Pterosaurs. In: Martill, D. M. and Naish, D. (eds.), Dinosaurs of the Isle of Wight, Palaeontological Association, Field Guide to Fossils 10, pp. 324-335.
• Seeley, H. G. 1901. Dragons of the air. Meuthuen and Co., London, United Kingdom, 239 pp.
• Wellnhofer, P. 1991. The Illustrated Encyclopaedia of Pterosaurs. Salamander Books Ltd., London. 192 pp.
• Witton, M. P. 2012. New insights into the skull of Istiodactylus latidens (Ornithocheiroidea, Pterodactyloidea). PLoS ONE, 7, e33170.