Monday 28 December 2020

Antique Paleoart: Dragons of the Air

Pterosaurs definitely do not get enough love, even these days. Of course, everyone knows who they are and there is rarely any dinosaur-media that leaves them out, but they are almost never the focus and are rather treated as an “Oh yeah, these existed too” or are even just there to add scenery. Once every blue moon, however, when the stars align and the magi sacrifice a virgin on Mount Ararat, it is decided that the pitiful reptiles will be granted a book of their own. The first instance of this happening comes to us from 1901 with Dragons of the Air: An Account of Extinct Flying Reptiles. This was written by none other than British paleontologist Harry Govier Seeley, who is maybe primarily known for being the one who split Dinosauria into the groups Saurischia and Ornithischia based on the shape of their pelvic bones. The book is rather technical and mainly filled with skeletal drawings and anatomical comparisons, but the few life-restorations in it reveal a remarkably prescient view for the time.

While Seeley does definitely classify pterosaurs as reptiles (and not as marsupials or weird bat-bird-hybrids as some have done before him), he nonetheless believed that they are the closest relatives of birds, renaming them Ornithosauria. Thus he reconstructed them with a similar physiology. This is rather remarkable, considering what other paleontologists at the time thought of them and how they would be stereotyped the following decades. Throughout the twentieth century pterosaurs were reconstructed as cold-blooded reptiles, which on the ground were only capable of awkwardly crawling or even resorting to sluggishly pushing themselves forward with their feet. Meanwhile Seeley here drew Cycnorhamphus suevicus as an upright quadruped, as would only become the standard for pterosaur-reconstructions almost 100 years later (complete with shrink-wrap!). In fact, the only difference between this reconstruction’s posture and that of modern ones is that the feet are digitigrade. Today we know, based off footprints, that they were plantigrade.

Most well-known of Seeley’s illustrations is likely his reconstruction of Dimorphodon, as it was used both in Peter Wellnhofer’s and Mark Witton’s books about flappy bois to show off how prescient he was for his time. Personally, I find the Cycnorhamphus above a bit more impressive, as the Dimorphodon here still looks vaguely bat-like. I assume Wellnhofer and Witton still liked to use Seeley’s Dimorphodon instead, as the quality of the drawing is much higher and more reminiscent of modern reconstructions. Perhaps it was also a way for them to stick it to Kevin Padian who throughout the 80s and 90s reconstructed Dimorphodon as a dinosaur-like biped.

If the latter is the case, Seeley does unfortunately also feature skeletal reconstructions of Dimorphodon in bipedal posture. He considers a bipedal posture for this animal plausible due to the shape of the fingers and claws, which he thinks look decidedly more raptorial than the feet and may have been used in seizing prey, again attesting to his dynamic, active view of these animals. Scaphognathus on the right is also reconstructed in bipedal posture, though notably next to a quadrupedal Rhamphorhynchus. It is probable that he made bipedal reconstructions in addition to the quadrupedal ones to draw a stronger comparison between birds and pterosaurs. While today’s consensus is that pterosaurs were quadrupeds, there still are some minority opinions, based off possible footprints, that some of them were bipedal. A 2020 study on such footprints from Korea has however revealed that most if not all of these supposed bipedal pterosaur-tracks were actually made by bipedal crocodylomorphs instead.

As we look at his Cycnorhamphus fraasii (which I believe would today be called Pterodactylus/Diopecephalus kochi) the most notable difference between this and modern reconstructions that comes to mind is of course the lack of pycnofibers, which we nowadays know pterosaurs definitely had (despite what Brazilian-fossil-poacher David Martill wants to tell you). This has been consensus since the 1970s, though Seeley does notably already discuss the idea of pterosaurs having feathers in this book. The reason is that as early as 1831 Georg August Goldfuss reported seeing filaments on the fossils of Scaphognathus and Germanodactylus. Unfortunately, Seeley and other paleontologists dismissed the idea and interpreted these filaments as wrinkles made by the wing-membrane, perhaps because the idea of fuzzy pterosaurs ran counter to the classification of them as reptiles. Seeley could have really done himself a favour by accepting Goldfuss’ interpretation, as it would have helped him in his major debates with Richard Owen, who viewed pterosaurs as cold-blooded primitives. In 2002 Goldfuss was eventually proven right as a study using UV-photography confirmed that what he saw were indeed fossil filaments.

Now here is a little lesson in science-history. What you are seeing here is a so-called quinarian system. This was a popular classification scheme among British zoologists before the idea of evolutionary family-trees became common. Remember that scene in the Dinosaurs TV show where Earl Sinclair classifies all dinosaur wisdom into the three categories animal, vegetable and rock? This is kinda similar. The base idea is that any one group of organisms can be divided into roughly five, or in this case six, sub-groups (and if you happen to find fewer than five it just means there are groups still waiting to be discovered). Nowadays this may seem highly simplistic, but it made sense at a time when most scientists still ascribed to the Aristotelian idea that most forms in nature are based off metaphysical templates. The diagram on the left is Seeley’s system of a group called Ornithomorpha, which he defines as the groups most adjacent to birds. As mentioned earlier, Ornithosauria (what he calls pterosaurs) he imagines as having developed in parallel to birds and is connected with them through the Saurischia and Aristosuchia (the latter being sort of a waste-basket taxon for the archosaurs he couldn’t shove into the other five groups). This is admittedly not too far off from how we see these relationships nowadays, though in other parts of the book he also classifies pterosaurs as the closest relatives of birds, both sharing a close common ancestor in the Paleozoic, making his views on relationships kind of confusing. For us it gets weirder once we view his larger-scale classification of the tetrapods on the right, with the Ornithomorpha being most closely adjacent to the Sauromorpha (a group he defines as comprising the turtles and tuatara, but also anomodonts, protorosaurs and plesiosaurs) and Mammalia. The latter he justifies through some mammal-like characteristics in the pterosaurs, comparing certain details of their anatomy with the synapsids found at the time in the Karoo Basin.

Lastly I just wanted to show you his small classification guide for the different pterosaur-groups because I think it is cute. Over all this was an interesting book to read through. It is really a shame that it took 90 years for another major book about pterosaurs to be released (The Illustrated Encyclopedia of Pterosaurs by Peter Wellnhofer) and after that one another 22 for Mark Witton’s Pterosaurs. The latter is still a great reference, though it has already aged in some areas through the past seven years. Pterosaur literature really ought to keep up with the current rate at which dinosaur books are published.

If you liked this and other articles, please consider supporting me on Patreon. I am thankful for any amount, even if it is just 1$, as it will help me at dedicating more time to this blog and related projects. Patrons also gain early access to the draft-versions of these posts.

Related Posts:

Literary Sources

  • Martyniuk, Matthew: Beasts of Antiquity. Stem-Birds in the Solnhofen Limestone, Vernon 2014.
  • Seeley, Harry Govier: Dragons of the Air. An account of extinct flying reptiles, London 1901.
  • Wellnhofer, Peter: The Illustrated Encyclopedia of Pterosaurs. An illustrated natural history ofthe flying reptiles of the Mesozoic Era, New York 1991.
  • Witton, Mark: Pterosaurs, New Jersey 2013.



  1. Ichnos published 'A catalog of pterosaur pedes for trackmaker identification' Peters 2011. There you will find several digitigrade pterosaur and pre-pterosaur (Rotodcactylus) tracks among the many others created by quadrupedal digitigrade pterosaurs. Seeley and Padian were correct with regard to Dimorphodon being a biped. Triassic Bergamodactylus could not have been a biped based on its proportions. Bennett also reconstructed Nyctodactylus as a biped supported by huge forelimbs, like ski poles with vestigial free fingers. Images here: Unfortunately Witton 2013 has several faults. Best to treat individual genera as individuals. There is great variety in pterosaurs. Here is Bergamodactylus:

    1. No offence David, but citing yourself as a source is a bit iffy, especially regarding someone of your reputation. Also, isn‘t a “biped supported by forelimbs” still technically a quadruped?

  2. Here is the 'catalog' at

    1. Yeah, that's a bit questionable to cite yourself.

  3. On books, don't forget David Unwin's 2005 Pterosaurs From Deep Time.

  4. "(despite what Brazilian-fossil-poacher David Martill wants to tell you)"
    Wait, what!? He seriously believes pyncofiberes aren't real, or am I missing something!?

    1. In a recent paper he suggested that some of the filaments preserved on an anurognathid are possibly actinofibrils instead of pycnofibers, but in the press-release he and Unwin went completely off the rails and claimed, despite this not being stated in their paper, that this might be the case for all pterosaur pycnofibers ever found.