Saturday, August 24, 2013

Pterosaur origins and the ‘protopterosaur’

One very interesting aspect of Mark’s new book is his take on the pterosaur ancestry issue. From decades people have reproduced an illustration by pterosaur researcher Rupert Wild that presented a hypothetical pterosaur ancestor. It was something rather lizard-like (since he considered pterosaurs closer to the prolacertiforms than current mainstream thinking of them as sister-taxon to the dinosauromorphs) but with elongating fingers and the beginnings of wing membranes. While often commented on that this probably wasn’t a great model anymore, it was pretty much the only nice illustration out there and so kept being re-used, and it did at least provide a conceptual idea of just what such an animal might look like. Lacking any transitional forms for pterosaurs, we do have to imagine a bit, and as I’ve long said, this is one area where palaeoart is key as it can communicate ideas like this to a lay audience in a way no amount of even florid prose can realistically do.

Mark’s approach however was not to produce a single transitional form but several of them. He worked his way through something very close to an archosaur so something nearly a pterosaur over three distinct stages. That gives a much more thorough take on this issue and allows him to explore not just what these things may have looked like, but also the evolutionary pressures and problems at different stages of their evolution. What’s interesting for me is that it also contrasts and compliments with something that I’ve been working on.

Most of you are probably aware that some years ago a major academic book was promised on the pterosaurs. The damn thing seems to be on near-permanent hiatus (I wrote my chapters back in 2008) but as part of the introduction I did on pterosaur origins I got together with Luis Rey and produced an updated ‘protopterosaur’ to take into account both the ornithodires, but also new data and hypotheses on wing evolution, basal pterosaurs and other issues. Luis has been kind enough to let me put up his effort and Mark has also passed on one of his intermediates for comparison (obviously these are copyright to these respective artists).
Marks' version of Rupert Wild's 'protopterosaur' (left) and his own suggested animal (right)
While obviously the art style is very different and we’ve got one image to covers everything (whereas Mark was working in stages) there are some strong similarities in places. The area of pterosaur origins came up in a chat between us perhaps two years ago and we soon discovered we’d both been producing protopterosaurs and that we’d independently struck on some strong connections. Sure this is all hypothetical, but it is interesting that we’ve headed down some similar lines in places starting from the same point and with the same assumptions.

For completeness and as a general guide, here’s the figure caption I produced for Luis’ art. I would have normally gone for a darker colour and less bright pattern to emphasise this as a small animal, vulnerable to predators, but when you ask Luis to illustrate and animal “not bold colours” should never be in your description. Still, the other bits of anatomy I wanted and the details are all in there and reflect various issues of pterosaur origins:
Luis Rey's version
Note that this is not being presented as a scientific hypothesis but is used to illustrate some of the profound changes that must have occurred to the ancestral species to become a pterosaur. This image was created in partial tribute to the oft reproduced ‘protopterosaur’ image by Wild (1984) which imagined a transitional form between prolacertiforms and pterosaurs. For this piece the ancestor was considered to be a dinosauromorph and with anurognathids or dimorphodontids as basal pterosaurs. The animal also includes new information on pterosaurs evolution (e.g. Bennett, 2008 on the origin of wing folding) to illustrate how some parts may have evolved. Different parts of this animal have been ‘allowed’ to evolve at different rate and thus some are much closer to the ornithodiran condition, and others to the pterosaur condition. Note the following features: the animal is arboreal; an archosauromorph-like head which is relatively short and tall, with and antorbital fenstra and archosaurian teeth; enlarged pterosaur-like orbit; head, neck, trunk and hind-limb proportions close to that of Scleromochlus; a relatively short and flexible tail exhibiting only a small tail-vane that is similar in shape to juvenile Rhamphorhynchus; an elongated wrist, but with no pteroid or propatagium; large manual digits 1-3 with large claws, a straight an elongate fourth finger that is still shorter than that of pterosaurs, this retains a small, anteriorly directed ungual; absent fifth finger; a relatively broad-chorded brachiopatagium that reaches only to the knee, but which is replete with actinofibrils; a simple archosaurian pes with an unmodified fifth toe and no webbing between the toes; a very limited uropatagium covering only parts of the legs and integrating with the base of the tail; only limited coverage of pycnofibers with much of the body still covered by scales.


  1. I guess the bright side of the stubborn refusal of transitional proto-pterosaurs to turn up is that your chapter won't need to be re-written when (if?) Pterosauria ever does come out. Amazon shows it as published in 2011 and now "Out of Print". Sold out before it was published! Think of the trees saved!

    1. Well I have had to keep updating things, and Chris Bennett has more stuff on the way so while 'finished', there is more to do I know. Still you're right, any major find now would entail and awful lot more work and as the chapter is already over 20,000 words I'm not desperate to have to really make any huge changes.

  2. It probably would be helpful to define or decide which of our known Triassic pterosaurs IS the most basal. That way you can find an outgroup genus that shares traits with it. Then it's imperative to find an outgroup for that genus and so on as Peters 2013 did in his Rio Pterosaur abstract and as Peters 2000 did covering the same topic. Notably absent from any discussion here are the Fenestrasaur candidates that currently share more traits with pterosaurs than any archosaur, especially Scleromochlus. That bipedal croc has a wide flat skull with a deep fossa surrounding its antorbital fenestra, doesn't have the right number of cervicals, has tiny hands unlikely to develop into wings, has no pedal digit 5, and has deep chevrons, unlike fenestrasaurs. With regard to the illustrations above, an attenuated tail is preferred for the ancestor of pterosaurs, as fenestrasaurs have and archosaurs do not. The addition of several vertebrae to the primitive sacral number in pterosaurs and fenestrasaurs points to a bipedal phase, as in Sharovipteryx and, by convergence, Scleromochlus. So why are these protopterosaurs clinging to tree trunks instead of perching on branches? BTW, if you supinate the hands, as Luis Rey did with digit 4, you have to also supinate digits 1-3 (palmar side anterior), which he did not do.

    1. Since you bothered to leave this comment I'll bother to point out how silly it is. I don't consider any Triassic pterosaurs likely basal (with the possible exception of Preondactylus which is only know from terrible / juvenile remains) and indeed nor do most researchers, preferring dimorphodontids and anuroganthids as I am sure you know.

      You also know full well that pretty much every part of your 2000 paper has been thoroughly taken apart in the literature and superceded by numerous other works in any case (Nesbitt's huge analysis most obviously) so no one accepts any part of it, and you know I do not so why whinge it's being overlooked.

      Citing a conference abstract from a conference that occurred 2 years after this picture was done is hardly something you can expect us to have taken account of, even if we thought it had any merit.

      Scleromochlus isn't a croc. You stating it is doesn't make it one. Pterosaurs are not 'Fenestrasaurs'. You stating it doesn't make it true.

      "So why are these protopterosaurs clinging to tree trunks instead of perching on branches?" Becuase as you clearly didn't bother to read, this is a deliberate tribute to Wild's artwork so is in the posture he gave it and for no other reason. As it happens I do think this is generally appropriate, but you are apparently trying to criticise something for a reason that not only isn't true but is explicitly explained. Well done you.

      You can continue to claim as much as you want that you have solved these issues and your answers are right and the rest of us are wrong, but simply putting in comments like this to claim we are ignorant or not listening is simply a waste of your time (and indeed mine to reply).

  3. Would love to see some evidence from you, David. Phylogenetic analysis usually carries a big stick, but Nesbitt didn't find any sisters that looked like pteros. Like you he avoided including the best candidates.

    Tribute or not, it's time for an update.