Feathery Fossil Gives Flying Dinosaurs a Size Boost

Early last week, in the pages of PNAS, paleontologist Dan Ksepka unveiled one of the largest dinosaurs ever to fly. With a 21 foot wingspan, the 25 million year old Pelagornis sandersi was a pseudotoothed, albatross-like bird that would dwarf any avian in the skies today, and even gives the previous record holder – the heftier Argentavis magnificens – some close competition. And now, in a case of fossil coincidence, paleontologists have announced a much, much older dinosaur that sets the size record for some the very earliest feathery fliers.

Named Changyuraptor yangi, this 125 million year old  dinosaur was not a bird. But it was close. Described by Bohai University paleontologist Gang Han and a team of experts in Nature Communications, this feathery creature was a non-avian dinosaur that took to the air in forests that once clothed Cretaceous China.

And while Changyuraptor shared many traits with close relative Microraptor, such as “hindwings” on the legs and elaborate tail feathers, this newly-named dinosaur was far bigger. While Microraptor was hawk-sized, Changyuraptor was about as big as an eagle, stretching about 4.3 feet from snout to tail tip. And on that big body, Changyuraptor had tail feathers stretching almost a foot long – the longest yet found on any non-avian dinosaur.

Despite all that plumage, though, Changyuraptor was not a bird. The very first birds split off from the rest of the dinosaur family tree about 25 million years earlier, and Changyuraptor, like Microraptor, belonged to a close, but non-bird lineage called dromaeosaurids. Nevertheless, Changyuraptor and kin may yield essential insights into how dinosaurs took to the air.

“There are two possibilities, which we still need to resolve,” says University of Southern California flight expert and study co-author Michael Habib. “This first possibility is that this tells us something about the evolution of flight because it’s a different way of flying than what leads to true birds,” Habib says, particularly because dinosaurs like Changyuraptor might represent an independent origin of flight that nevertheless shared some common facets with what early birds were doing. This sort of evolutionary convergence might indicate shared constraints on how feathered dinosaurs could evolve flight.

“The other possibility, which is better-supported at present,” Habib says, “is that these four-winger kinds of morphology were actually quite common amongst paravians”, or the larger group that includes birds as well as their close, feathery, non-avian relatives. The fact that many dinosaurs near the base of the bird family tree – either on that line or close to it – had elongated leg feathers means that dinosaurs like Changyuraptor can act as proxies for how birds evolved flight.

In either scenario, Habib notes, Changyuraptor and other quad-winged dinosaurs suggest that “protobirds would have a lot of different control surfaces.” Rather than the long leg and tail feathers being oddities indicative of a weird way of flying, these structures may have been crucial for flying dinosaur stunts in the days before the majority of flight control shifted to the front wings. Changyuraptor doesn’t represent the addition of extra wings, but may embody critical surfaces that the first birds and other flying dinosaurs needed to turn, brake, and land early in their evolution.

But despite the clear hallmarks of aerodynamic adaption in the bones and plumage of Changyuraptor, “We don’t know how it was actually moving through the air,” Habib says. As yet, paleontologists are still narrowing down possibilities.

Soon after the discovery of Microraptor, for example, some paleontologists reconstructed the dinosaur flying with a spread-eagle or biplane posture. But this doesn’t make anatomical sense.

“Until we find a microraptorine that shows anything else, I wouldn’t expect them to have much more flexibility at the hip than other theropods,” Habib says, meaning that Microraptor and Changyuraptor probably held their legs directly beneath their bodies while moving through the air. And even if these dinosaurs had a little more hip flexibility than usual, Habib adds, that could be an adaptation to tree-climbing or some other mode of locomotion on a surface rather than flight. Changyuraptor was not the dinosaur equivalent of a flying squirrel.

Rather than acting like a modern bird or like an earth-bound Velociraptor, Changyuraptor probably had unique habits that were lost to Cretaceous time. If forced to sketch the dinosaur’s lifestyle, Habib says, he’d expect that “these things are semi-arboreal, landed on the ground but like to be in trees, and probably could execute dynamic and relatively short aerial maneuvers” like turning and braking fast to get to the ground, perhaps to pin down small prey. Maybe they were even able to flap their wings a little to give themselves a little more push, Habib says, but even within such scenarios, the extensive feathering of dinosaurs such as Changyuraptor indicates that once these creatures got into the air, they were able to use their arms, legs, and tails to deftly navigate through the dense stands of trees they flew through.

Understanding how Changyuraptor flew, and what the dinosaur means in the broader picture of how flight evolved, relies on future research. No paper stands alone. Habib is already on it. “One nice thing about doing the analysis with this animal, developing predictions of its tail performance, is that it’s probably applicable” to other feathered dinosaurs, Habib says. “The interest in the animal – new species, big, nice tail – is that it’s a springboard for a broader analysis of paravian tails.” Putting all those pieces together, plumage and prehistoric bone alike, connect toothy little raptors to the chickadees hopping about outside your window.

Reference:

Han, G., Chiappe, L., Ji, S., Habib, M., Turner, A., Chinsamy, A., Liu, X., Han, L. 2014. A new raptorial dinosaur with exceptionally long feathering provides insights into dromaeosaurid flight performance. Nature Communications. doi: 10.1038/ncomms5382