How pterosaurs became airborne
While it’s easy to generalise birds, bats and pterosaurs as the result of very lucky gliders, the truth is that we know very little about how these animals became airborne. Aerial dinosaurs are notoriously subjected to intense debate, especially considering that forms like Microraptor are amazingly highly unsuited for both climbing and WAIR, but birds are hardly the only mysteries. Bats, often credited as derived from gliding mammals, might actually be the result of a highly unusual cave dwelling lifestyle, though predictably this is highly criticised as well.
In both birds and bats, therefore, can can negate origins from tree climbing gliding ancestors, the usual models for flying animal ancestors. So, in that light, how did pterosaurs, the other group of flying vertebrates, became airborne?
What we currently know
More controversial than any other thing about them, is where pterosaurs fit in the tree of life. The most valid suggestion, which is currently the “official” one, is that pterosaurs evolved from archosaurs closely related to dinosaurs, and that both Pterosauromorphs and Dinosauromorphs form clade Ornithodira, or “bird necks”.
And within Ornithodira, pterosaurs are often considered to be closely related to the archosaur Scleromochlus, a little sauropsid that is pretty much the archosaurian answer to Sharovipteryx. Even then, this is not certain; some consider Scleromochlus to be a more “primitive” ornithodiran, or even outside of the clade all together. The fact that it has long hindlimbs and short forelimbs, as opposed to the long forelimbs and short hindlimbs of pterosaurs, doesn’t help matters, though it’s worth to note that gliding squirrels do have more developed forelimbs than other squirrels, so such a reversal is not unheard off.
Within Pterosauria itself, the most basal taxa don’t help matters. Currently, the most basal known pterosaurs are anurognathids. Anurognathids were highly different from the stereotypical pterosaur: with very short and wide toad-like jaws, massive eyes and short, broad wings covered in specialised pycnofibers, these were a far call from the elegant, diurnal, colourful pterodactyloids of the Cretaceous, being basically the closest a sauropsid ever came close to a bat. More importantly, however, they provide clues as to how pterosaurs evolved.
Anurognathid morphology is consistent with that of arboreal, nocturnal/crepuscular insectivores. Their short and wide snouts were highly remaniscient of not just amphibian mouths, but also of the beaks of nightjars, swifts and swallows, all acrobatic aerial predators of moths and other insects. Their wings were proportionally short by pterosaur standards, being quite adequate for flying among dense forests, and the wing membranes were bordered by long pycnofibers similar to the barbs of silent-flight birds like owls, indicating that these animals had quite silent wing beats. It’s very likely that they were ambush predators, relying on the cover of darkness and their silent wings to stalk their prey undetected, and likely to avoid potential predators like arboreal mammals and, latter on, larger pterosaurs. We also know that they had unique crests on their arm bones, implying that these were also among the few known pterosaurs to be capable at hoovering.
Like all non-pterodactyloid pterosaurs, anurognathids were also highly adapted to climbing. Their claws were robust and curved, impairing movement on flat surfaces, but allowing efficient climbing. Of what we know about their wing membranes, they were more extensive on the legs than in more derived pterosaurs, also making running cumbersome. These were, therefore, highly arboreal animals, presumably avoiding the ground as much as possible.
Therefore, based on Anurognathidae, we can speculate that the first pterosaurs were nocturnal, arboreal gliders/flutterers. The tree tops of the Triassic had a menagerie of diurnal arboreal reptiles like drepanosaurs, and back then there was virtually no competition for the numbers of flying insects. With competition on the ground and availiable prey on the air, flight was an inevitability.
There are modern analogues!?
Most surprising, is that there is an echo of the old ornithodires that produced pterosaurs, in the dense tropical rainforests of the Holocene.
Gliding or flying geckos are curious squamates distributed across four genera. The most famous, and ostensibly the most aerial, are the Ptychozoon geckos from Southeast Asia, but there is also the closely related Luperosaurus and the distantly related Thecadactylus.
All these gekkotans are nocturnal, arboreal animals just like most geckos, but thanks to the extensive flaps of skin that stretch along the limbs, in the flanks and between the toes, they can glide efficiently. As said before, Ptychozoon, with the most extensive membranes and wider tail, is the most aerial of these geckos; they are frequently captured in bat nets, suggesting that they engage in aerial locomotion frequently, and perhaps even to capture aerial prey.
Most interestingly, unlike other gliding squamates, which rely on ribs to support the wing, the flank membranes of these geckos are unsupported. Indeed, these animals rely a lot on the webbed toes to form the wing surface; such a reliance on the limbs, combined with already present flank membranes, makes the evolution of powered flight the logical conclusion, and the ancestors of pterosaurs likely followed the exact same path, especially when marks of membranes between the pterosaur clawed fingers have been found.
The evolution of “new pterosaurs” from Ptychozoon is probably sadly postponed, as bats and nocturnal birds rule the skies of their environment. However, long ago, in the Triassic, there were no flying dinosaurs or mammals, and as such animals very similar to Ptychozoon were free to further develop their wings.
(EDIT: Before proceeding with the comments’ section, read Darren Naish’s post on the phylogenetic controversies raised by David Peters)