The image of dinosaurs has changed a lot over the years. After a brief period of being considered active, agile creatures, for most of the 20th century they were relegated to tail-dragging swamp creatures. Through the 80s and 90s, new evidence and new interpretations of old evidence led to a change in how dinosaurs were perceived. This new look for dinosaurs was solidified in the public eye by Jurassic Park. It was only a few years later that dinosaur fossils were found with traces of actual feathers. Since then, there has been so much more evidence uncovered that it’s hard to figure out which dinosaurs probably had feathers and which didn’t. Whether you’re an artist looking for inspiration or just somebody curious about the topic, I’m trying to condense everything currently known in a digestible format.
The best place to start is the start; when exactly did dinosaurs evolve feathers? This matters a lot, because if they evolved them early, before they split into multiple branches, every branch could potentially have feathers. If it was evolved late, only the one branch that evolved it would have feathers. At first, the evidence leaned towards the latter, with all known feathered dinosaurs being members of the coelurosauria, the group of small predatory dinosaurs that includes Velociraptor, Archaeopteryx, and modern birds. However, with the discovery of quills on Psittacosaurus (a small, bipedal relative of Triceratops, on the opposite side of the dinosaur family tree) and the suggestion that dinosaur feathers and the fuzz on pterosaurs was actually homologous (meaning it evolved from the same origin instead of developing separately), it is now clear that feathers developed long before dinosaurs split into their various lineages. Any dinosaur could have had feathers.
But which dinosaurs did? There’s dinosaurs which had feathery ghosts surrounding their bones when we found them, but trace fossils like that are rare. Most of the time, it’s a case of inference, often through relationships. There are no fossils of Velociraptor with feathers, but their relative Microraptor has full wings, and Velociraptor arm bones have bumps that appear to be quill knobs, attachment points for large wing feathers. From this, it can be safely inferred that Velociraptor and its other relatives had feathers, including full wings, although not flight-capable ones.
These full wings were rare among dinosaurs, and only the most birdlike among them had them. This included the dromeosaurs (Velociraptor and its relatives) the troodontids (similar to dromeosaurs but even more gracile and bird-like) and the Oviraptors (similar to the others in build, but with short, parrot-like beaks instead of toothy jaws). The ornithomimosaurs were the dinosaurs most distant from proper birds to be outfitted with full wings. Most famous for Galimimus, which made a minor appearance in Jurassic Park, these dinosaurs were shaped like a modern-day ostrich or emu but for their long tails and functional arms.
What did these wings look like, and what were they for if not flight? In modern birds, the wing feathers are attached to the second finger; this isn’t so obvious because their fingers are fused into a simple lump of bone, but in these dinosaurs, it would have been extremely clear. The first finger was clear of the wing, while long feathers would cover the remaining fingers, although in many of them the wing would be short enough that a claw or two would still be visible from the side. In all groups, the wings were short and broad, like a downscaled chicken wing, and could have had a few purposes.
Modern bird chicks, long before they can fly, will flap their wings when running up steep slopes, gaining extra traction. This could also have been used for balance while grappling prey, a function modern eagles use their own wings for. The least flashy use of wings is one that some non-avian dinosaurs definitely had for them. The skeleton of an Oviraptor was found over its own eggs, arms outstretched. Wings would have trapped the raptor’s body heat, allowing it to incubate its eggs without putting its full body weight on them. It’s a trick that many modern birds use as well.
The group of dinosaurs known to have wings is called the Maniraptora. Outside of them, the coelurosaurs still had feathers, but only the most basic kind, a simple downy covering. Like the hair of modern mammals, this could have varied greatly in length from species to species, with polar creatures having longer hair and those closer to the equator having shorter hair (or none, we’ll get to that). While most coelurosaurs were small and could use the boost to their heat retention, several groups did get massive, and one of those was the Tyrannosauria. Whether Tyrannosaurus had feathers has been a contentious topic that has gone back and forth ever since it was realized that they were widespread among dinosaurs, but what’s the current evidence? Was it feathered like its relatives, or was it just too big?
It is a relative of Tyrannosaurus that tells us that even large dinosaurs could have a coating of feathers. Yutyrannus huali was roughly as large as an Allosaurus fragilis, both being about 30 feet long. Unlike Allosaurus (as far as we know), Yutyrannus was completely covered in a coat of feathers. However, there are two main differences between Yutyrannus and Tyrannosaurus, besides their relative distance on the tyrannosaur family tree. For one, as big as Yutyrannus was, Tyrannosaurus was much larger. A large adult was ten feet longer than Yutyrannus and built far more heavily.
Bigger bodies produce more heat and shed it less efficiently, due to the fact that volume increases faster than surface area when size increases. There’s more inside producing heat and less outside proportionally shedding it. Second, Tyrannosaurus lived in a warmer habitat. It would rarely have seen snow in most of its range. With this combination of magnitude and a warm environment, it seems likely that Tyrannosaurus was the elephant to Yutyrannus‘ mammoth, with sparser feathers, if any. The current evidence leans towards, this, with all current skin impressions from Tyrannosaurus showing either scales or bare skin (meaning no scales or feathers, something that many modern birds have in places and many other dinosaurs probably had somewhere on their bodies). Tyrannosaurus might have had small patches of feathers for display, no feathers at all, or a thin coating like a modern elephant’s hair.
Outside of the coelurosaurs, there is a problem. Because the earliest dinosaurs, all lineages of dinosaur had the potential for them. However, most preserved skin remnants show only scales. What this suggests is that scales were lost early on in the development of most dinosaur groups, possibly when they began to grow larger. Our evidence of large dinosaurs in most clades carrying feathers is limited to conjecture. For example, among what used to be called the carnosaurs, your standard large theropods most famously known for members like Allosaurus, the best evidence comes from an oddball called Concavenator.
Concavenator was a relative of some of the largest theropods of all time, like Acrocanthosaurus and Charcharodontosaurus, but was quite small itself, standing apart more for its unusual anatomy than its unusual size. A small hump or sail was centered entirely over their hips, a strange little bump like a shark’s fin. It also had bumps on the bones of its forearms that resembled the quill knobs on the bones of modern birds and Velociraptor. If they really did serve a similar purpose, Concavenator could have had long feathers on its arms, likely for display. However, some paleontologists have suggested that these quill knobs were actually regular muscle attachments, and there still isn’t any clear evidence either way of feathers in the carnosaurs.
Similarly, on the other side of the dinosaur family tree, feathers seem mostly to have been present in smaller, more primitive forms and lost in the larger, advanced giants later in their lineages. Kulindadromeus is a very standard basal Ornithischian, a small herbivorous biped with a long tail and short front legs. Similar dinosaurs appear throughout multiple lineages, and until recently they were all assumed to be completely scaled. The only unusual thing about Kulindadromeus is that one specimen has preserved integument, showing that it had feathers on most of its body, but scales on parts of its tail and legs. This provides an unusual bit of insight into dinosaur life appearance that should be kept in mind when reconstructing them; feathered dinosaurs could still have scales on parts of their bodies in interesting patterns, and mostly scaled dinosaurs might still have some feathers.
Proving this second option further is Psittacosaurus. Psittacosaurus was an early relative of Triceratops and its horned kin, although you wouldn’t know it just looking. Psittocosaurus didn’t have horns or a bony frill, and it could have walked on only two legs if it wanted to. Detailed skin impressions show that most of its body was covered in scales, but the back of its tail grew long quills. These would have been for display, and they open the possibility of all sorts of display feathers all across dinosauria. Some artists have even interpreted later ceratopsians like Triceratops itself with similar quills, a neat idea that is probably inaccurate (Triceratops had its horns and frill as a display space, and what have been interpreted as attachment points for quills have better explanations) but a great inspiration.
Everything here is subject to future discoveries, of course. The day after I upload this article, an allosaur might be found with a complete covering of feathers, or a raptor that become secondarily scaled. These are just the general patterns we find across the dinosauria for body coverings. If you’re wondering about scales, they have just as much diversity, too much to go into here. Every group of dinosaur was unique in a thousand ways, and the covering of their skin was part of that.
Want more on dinosaurs and other prehistoric life? Check out the links below!