Monday 16 June 2014

Darwinopterus vs. Bat Out of Hell

Darwinopterus robustodens and his pal, a European robin (Erithacus rubecula). When not posing for artwork, they drive around in a van solving mysteries.
Wukongopterid pterosaurs have been on my mind this week thanks to a near-complete manuscript about them. This required* rendering everyone's favourite wukongopterid Darwinopterus - results above (note that this is D. robustodens, not the more familiar modularis). In the final version he's joined by something else, but you'll have to wait to just what that is. Wukongopteridae is the group of recently discovered Chinese pterosaurs which bridge early pterosaur and pterodactyloid-grades of pterosaur morphology, famously combining head and neck characteristics of the latter with the bodies of the former. They're best known for Darwinopterus modularis, but there's actually now nine(!) taxa known from the same horizon in north-east China, all morphologically very similar and almost certainly oversplit.

*By 'required', what I really mean is that I'm compelled to construct papers with extraneous artwork in them somewhere, because I'm a sucker for punishment.

Illustrations of Darwinopterus and I go way back. I was asked to do press images for two early publications on it: the initial description and assessment of its unusually 'modular' evolution (Lü et al., 2010), and the equally fantastic discovery of its sexual dimorphism and an egg-mother association (Lü et al., 2011a). Below is the first of these images, published in 2009.

Ah, 2009. Shrink wrapping, relatively light integuments and very low soft-tissue crests were still in fashion. Fun fact: Darwinopterus was referred to as 'Frank' before it was given a binomial, a nod to it being a Frankenstein's Creature-like mash of body parts.
Looking back on Darwinopterus 2009, I'm not enormously happy with it. This in itself isn't that unusual. Artists often look back with dissatisfaction with older works, but this has an additional reason for dissatisfaction: I never really agreed with the notion of Darwinopterus as an aerial-hawker of flying tetrapods, and I think this comes across in its execution. I outlined my basic concerns with this idea in Pterosaurs (Witton 2013):
Given that wukongopterids have provided an insight into macroevolutionary processes, filled a gap in pterosaur phylogeny, and present a very unique pterosaur bauplan, expectations may be high that their proposed foraging strategies will also be rather amazing. Fittingly, some have proposed that wukongopterids were pterosaur top guns, their newly evolved long necks and oversize heads being used to prey upon dinosaurs, other pterosaurs, and even gliding mammals in midair (Lü et al. 2010). Such acts would be rather remarkable because, with even a generous mass estimate, the biggest Tiaojishan wukongopterids would not weigh much over 300 g (extrapolating data from Witton 2008), which is about the same as a modern feral pigeon.
At that size, tackling squirrel- sized mammals or crow-sized dinosaurs on the wing would be a feat earning praise from even the hardiest modern raptors, and wukongopterid skeletons would have to be brimming with offensive weaponry for this purpose. Vertebrate-hawking birds are renowned for their talons, incredibly strong feet, robust skulls, and powerful beaks (e.g., Hertel 1995; Fowler et al. 2009), while bats that subdue large vertebrates in flight are also armed with formidable teeth and powerful jaws (Ibáñez et al. 2001). These adaptations provide the means to immobilize their prey quickly and efficiently, and are obvious advantages for animals grappling with large prey while in “flight. Vertebrate hawkers are also powerful fliers that can chase down their quarry and, once immobilized, carry the prey to a safe spot to eat. Pterosaurian equivalents would therefore require equally powerful “flight musculature to permit the same tasks. Unfortunately for the Darwinopterus raptor hypothesis, wukongopterids do not possess any of these requirements. None of their appendages bear the chunky digits and talons ideal for subduing large aerial prey items, and their long, comparatively delicate skulls and unimpressive teeth are ill suited to this task. Nor, for that matter, do they have the expanded shoulder regions indicative of the powerful “flight muscles needed to chase and eventually carry their prey. With this in mind, raptorial pursuits look doubtful for wukongopterids.
Witton (2013), p. 141-142.
There's a lot more that could have been said about this, but you get the idea: Darwinopterus and chums were small (jackdaw-sized - see image at the top of the post), delicately built animals for which aerial predation seems counter-intuitive and unlikely. I'm not the only person saying this, either: Lü et al. (2011b) and Sullivan et al. (2014) also raise points against the aerial hawking idea. We could go so far as to to label wukongopterid aerial hawking as another example of an 'extreme' palaeoecology based on cherry picked characteristics rather than considering a full suite of functional data.

So, on reflection, we probably started on the wrong foot but, hey, what can you do? Working as an artist is quite different to being involved in research: ultimately, you're a guy with a paintbrush being told to illustrate someone else's idea, even if you don't necessarily agree with it (also see Csotonyi and White 2014). With the notion of aerial predation being raised in the paper, specifically as a possible explanation for the development of pterodactyloid head and neck features Lü et al. (2010), it was an obvious choice for the image. But we didn't help matters by walking into some of the cheesiest, silliest things stereotypes of palaeoart, primarily because we were trying to make a small, fairly inoffensive animal look like a skydiving, dinosaur-eating-badass. We tried several different takes on this, varying aerial and terrestrial prey, and compositions which upped the voracity, such as this:

You'll need to make your own whooshing jet fighter noises and 'pew pew' laser sounds.
Eventually death-from-above was decided over death from sideways. But a key issue to tackle was that that the proposed prey for Darwinopterus was the same size as the predator itself (Sullivan et al. 2014 go into this more), and it's obvious that Darwinopterus is not a fighting, wrestling creature. To make the image work at all we had to play liberally with animal sizes: the maniraptoran (loosely based on Anchiornis in the final version) is tiny, and the Darwinopterus is a huge, ferocious juggernaut. I'm not the only artist who used this trick: the disappointing pterosaur documentary Flying Monsters 3D also had their Darwinopterus plundering undersized theropods. After we tinkered with reality, we then started piling on the cheese: the maniraptoran was made more reptilian-looking, it's head twisted to stare into the Maw of Destiny and, of course, it's mouth open to 'NOOOOOO!!!' it's impending fate. The result is 'Meat Loaf palaeoart', the sort which resembles the worst kind of rock album covers more than nature.

Think I'm being daft comparing palaeoart to cheesy hard rock album covers? How many depictions of extinct animals have the exact same pose as the titular bat here? See links below for more examples. From Wikipedia.

The result isn't awful, but I think I've had more successful collaborations with the same authors, just because this doesn't feel realistic at all. From the science to the composition, the whole thing just seems 'forced'. Don't get me wrong: I'm sure Darwinopterus could be terrifying instruments of mortality to the right prey (probably terrestrial invertebrates according to Lü et al. 2011b and Witton 2013), and I'm not belittling the drama or viciousness which can occur in the lives of small animals. But life - thank God - doesn't look like the cover of a Meat Loaf album. There are some fossil animals that can - just about - pull off Meat Loaf palaeoart, but it's just not possible to turn small, fluffy, unthreatening animals into monsters. When we do, they're more liable to look goofy than impressive. Despite this, we do it all the time. I understand why we do - it's exciting and marketable, and feeds into expectation that Deep Time as a monster-filled fantasy realm - but it also reinforces the perception that palaeoart is unsophisticated and aimed at children. But there may be another way.

Palaeontologists: next time you have a new, diminutive animal you want illustrated, forget the monster: play the cute card. The chubby little Darwinopterus at the top of this post intuitively seems closer to the reality of this animal than Darwinopterus 2009: AwesomeoSuperKiller. It's proportions, fluffiness and posture are all accurate, and it only looks cute because, well, it probably was: it's a small fuzzball with an oversize head and mischievous grin. If the Internet's obsession with cats and baby sloths has taught us anything, it's that cute sells better than violence. Indeed, this image is one of the most popular things I've ever put on Facebook, and (at time of writing) it was posted less than a day ago. The great thing about the cute card is that everyone wins: the artwork should promote research just as well or better than it's Meat Loaf variant, because it appeals to wider demographics. The artist gets to render an animal in a more realistic light without jumping through hoops to monsterise it, and the world gets a new picture to 'coo' at. A few folks might even start to appreciate extinct animals in ways they never could when they're always shown screaming and fighting.

And if nothing else, it will mean we'll never have to discuss Meat Loaf album covers here again.

May God have mercy on us all. (Wikipedia)


  • Csotonyi, J. & White, S. (2014). The Paleoart of Julius Csotonyi: Dinosaurs, Sabre Tooths and Beyond. Titan Books, London.
  • Fowler, D. W., Freedman, E. A., & Scannella, J. B. (2009). Predatory functional morphology in raptors: interdigital variation in talon size is related to prey restraint and immobilisation technique. PloS one, 4(11), e7999.
  • Hertel, F. (1995). Ecomorphological indicators of feeding behavior in recent and fossil raptors. Auk, 112(4), 890-903.
  • Ibáñez, C., Juste, J., García-Mudarra, J. L., & Agirre-Mendi, P. T. (2001). Bat predation on nocturnally migrating birds. Proceedings of the National Academy of Sciences, 98(17), 9700-9702.
  • Lü, J., Unwin, D. M., 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: Biological Sciences, 277(1680), 383-389.
  • Lü, J., Unwin, D. M., Deeming, D. C., Jin, X., Liu, Y., & Ji, Q. (2011a). An egg-adult association, gender, and reproduction in pterosaurs. Science, 331(6015), 321-324.
  • Lü, J., Xu, L., Chang, H., & Zhang, X. (2011b). A new darwinopterid pterosaur from the Middle Jurassic of western Liaoning, northeastern China and its ecological implications. Acta Geologica Sinica‐English Edition, 85(3), 507-514.
  • Sullivan, C., Wang, Y., Hone, D. W., Wang, Y., Xu, X., & Zhang, F. (2014). The vertebrates of the Jurassic Daohugou Biota of northeastern China. Journal of Vertebrate Paleontology, 34(2), 243-280.
  • Witton, M. P. (2008). A new approach to determining pterosaur body mass and its implications for pterosaur flight. Zitteliana, 143-158.
  • Witton, M. P. (2013). Pterosaurs: Natural History, Evolution, Anatomy. Princeton University Press.


  1. Mike from Ottawa16 June 2014 at 10:02

    I'm wondering why anyone would think a long neck an adaptation for aerial hawking on vertebrates. It doesn't seem to be a feature of any of the extant members of that guild that I'm aware of, but then I'm not aware of much, so are there any?

  2. Mike from Ottawa16 June 2014 at 10:12

    I should add that it's an interesting post. Always fun to learn about the science-art interaction and how views of extinct animals change with reflection and new information. Finding out how we know is the best thing about internet outreach by scientists and is something that doesn't get as much play in other media. That the post engaged me is shown by my instantly going to a question on substance rather than kudos on the post, and the 'new' Darwinopterus.

    1. Spot on with the neck thing - I actually mentioned this as an addendum to my book text in one draft of this essay. There's two issues here. Firstly, the long neck and head would move the centre of gravity well forward tremendously when carrying large prey, making for strenuous and ineffective flying once prey was caught. Reason two: having a long neck and head is really not ideal for catching prey on the wing anyway, given the strain involved. Careering into another large animal, flying or not, at high speed with your long, slender mouth agape atop your long, slender neck bones is something most wukongopterids could only do once. This is why avivores and hawking bats have much shorter, chunkier skeletons up front.

  3. Mark I think you should develop a painting out of that pencil sketch, I love that pose, great composition.

    1. Thanks Stevie, but I doubt it'll happen. I just don't see these sort of habits really reflecting what we know about this animal, so I'm not really interested in taking the idea further. It might look cool, but I'd prefer to render something both cool _and_ feasible.

    2. Adjusting the composition so that it's chasing a dragonfly or similar might work both scientifically and artistically.

    3. Not a bad idea, although I don't really see these guys as aerial insect chasers either. There's a lot of convergence in aerial insect hawkers - short, wide skulls, lots of wing modifications for slow flight - and wukongopterids don't show any sign of these. Seems more likely that this thing was a pterosaurian blackbird: hopping about on the ground, picking up small insects and the like - rendering that is of much more interest to me than continuing the poorly-supported aerial hawking trope.

  4. The most misleading thing about the ferocious Meatloaf album artwork is, well, the music is not ferocious at all.

    1. Mike from Ottawa16 June 2014 at 21:14

      In fairness, the music is as ferocious as your average meatloaf, so folk were warned.

  5. I agree with Steve, maybe the small maniraptoran should leave the picture and instead could be replaced by an competitor for territory and females...

    Great post!

    All the best,

    Joschua Knüppe

  6. Oh oh, I get the joke.... nananananananananananana... bat from hell and robin!

    *hides behind rock*
    sorry, couldn't resist. 'Love the art though.
    Out of curiosity, (and because the extra fluff made your Darwinopterus look more natural) why do you still prefer to draw Azhdarchids with relatively slim necks? Is there anything about the neck part of the skeleton that leads to a flamingo like restoration (re: thin, not crazy s-shaped)?

    1. "Is there anything about the neck part of the skeleton that leads to a flamingo like restoration (re: thin, not crazy s-shaped)?"

      There is indeed. Firstly, it's not so much that azhdarchid necks are thin, but instead incredibly long: 4-5 times body length in some species. That naturally stretches the soft-tissues out over a long distance, making them appear thinner than they are. On top of this, azhdarchid mid-series cervical vertebrae are famously tubular, lacking the large processes and complex shapes which anchor big neck muscles, and there's fairly limited motion between each bone. These are not robust, powerful necks, but instead relatively immobile structures always held in gentle arcs: that doesn't require a huge amount of muscle power. Finally, because most azhdarchids are big, they can't really play the 'fluff' card for beefing out their neck proportions: they'd need some seriously long and dense pycnofibres to bulk them out. Little species (like this Darwinopterus) are going to look poofy with even 'standard' levels of fluffiness.

    2. Thanks! *considers self properly enlightened*, huge flying pick axes...

      Why were they growing long necks anyway? It seems they were just really there to hold the heads up, and not launch them (the head) at prey/food items (and I don't mean in a killstorkodoom vs. captain deathbeast way)

      Oh, wait, 'raining on Darwinopterus' parade. 'Guess I'll just wait for an azhdarchid post. These are forthcoming... right?