Feathered Dinosaurs: How do we know?

For a long time feathered dinosaurs just looked weird to me. Seeing fuzzy Deinonychus or some other dromeosaur with a splash of plumage never looked quite right and I didn't understand why in the course of a few years predatory dinosaurs went from being scaly to being covered in down. Most of the books I had seen didn't explain it beyond "These dinosaurs were closely related to birds," something I didn't dispute but was not enough to make me feel comfortable with feathered raptors. Even after I started taking a greater interest in paleontology I still had problems with reconstructions of feathered dromeosaurs when no specimens with feathers had yet been found. As I learned more, however, I came around to the notion that any Deinonychus or Velociraptor that appears without feathers is a naked dinosaur, a growing body of evidence showing that there were probably many more feathered dinosaurs than had previously been suspected.

The notion that birds and dinosaurs are closely related has been around for a very long time, kicked off by the discovery of Archaeopteryx and Compsognathus in the mid-19th century and popularized by T.H. Huxley. (It should be noted, though, that similarity was not always considered to reflect an ancestor-descendant relationship.) Even though the term "dinosaur" had only just been coined by Richard Owen in 1842 by the 1870's the Solnhofen fossils, trackways, and the bipedal Hadrosaurus and Dryptosaurus spurred the first Dinosaur Renaissance, revealing bipedal (and even bird-like) animals rather than the pachyderm-like creatures Owen brought to life at the Crystal Palace. The smaller dinosaurs like Compsognathus and Hypsilophodon were particularly important as they were considered to be more representative of the form of bird ancestors, flightless birds like rheas and emus being the next step in the hypothetical evolutionary system. Still, it wasn't until the second Dinosaur Renaissance of the late 20th century that the notion of birds as living dinosaurs began to hatch.

A particular problem plagued the hypothesis that birds evolved from dinosaurs, however; no feathered dinosaurs had been found. Archaeopteryx was considered to be a bird, too derived to be comfortably called a feathered dinosaur, and while it became increasingly important as a transitional fossil a transition from what was debated. Eventually the morphological problems that faced the dinosaur hypothesis, like the supposed absence of clavicles, were overcome with new evidence and increased study but more than Archaeopteryx was needed to confirm the predictions being made. Enter Sinosauropteryx prima, described in 1996 (Chen et al 1998). Although there was some skepticism about whether or not the preserved structures were really feathers (Unwin 1998, Thomas & Garner 1998) a flood of feathered dinosaurs coming out of China soon followed and overwhelmingly supported the notion that birds evolved from dinosaurs.

Specimens like "Dave" (a probable Sinornithosaurus, Ji et al 2001, the first specimen of Sinornithosaurus millenii also bearing filamentous feathers [Xu et al 1999]) threw greater weight to the notion that birds had evolved from predatory dinosaurs. (For a fairly recent reviews see Norell & Xu 2005 and Zhou 2004.) Even more surprising, however, was the discovery of Beipiaosaurus inexpectus, a therizinosauroid dinosaur with integumentary feathers. While they are theropods relatively closely related to the raptors therizinosauroid dinosaurs are also very different, having long necks, huge claws on their hands, and perhaps a herbivorous diet. If these dinosaurs had feathers it raises the question of whether other coelurosaurs, which include the ostrich-like ornithomimosaurs and the tyrannosaurs, also had feathers. It is entirely possible that even the terrifying Tyrannosaurus had plumage during at least some stage of its life, perhaps to regulate body temperature when small but shedding the feathers as it quickly became larger, and even the recently described Jurassic tyrannosauroid Guanlong (Xu et al 2006) may have had feathers. This hypothesis has yet to be confirmed but it may be that feathers are not just indicative of dromeosaurs but of coelurosaurs as a group.

Feathers pop up again and again in differing forms and arrangements in coelurosaurs, and if we think about this evolutionarily the prediction that at least all dromeosaurs should be feathered is clear. If feathers were not inherited from a common ancestor then they would have to evolve several times within the coelurosaurs, such large-scale convergence being unlikely. If different groups of coelurosaurs were feathered it is reasonable to assume that most, if not all, were, and any that lacked feathers would have secondarily lost them for one reason or another. This prediction is still being worked out but future studies may help confirm or refute it, especially since there is more than one way to detect the presence of feathers in the fossil record.

The discovery that truly blew me away was revealed in a short note printed in Science last year. Until recently I had assumed that the detection of feathered dinosaurs relied entirely upon exceptional preservation, feathers being so delicate that in most cases they will not be preserved (indeed, Archaeopteryx specimens in which the feathers were not well-preserved were mistaken for Compsognathus and pterodactyls). If the feathers are not preserved we can still make the case that the dinosaur would have been feathered based upon its relationships but confirming the idea would require an exceptionally-preserved specimen, the chances of finding one becoming increasingly less likely with increasing body size. What Turner et al (2007) found, however, were quill knobs on the forearm of Velociraptor, the very same structures seen on living birds with secondary feathers. The terrors of Jurassic Park had more than the sparse mohawks they sported in the third installment; they had what would appear to be flight feathers, perhaps being used for display since they certainly could not fly.

Regardless of what they were used for there is now an osteological character that can be observed to see if some dinosaurs had secondary feathers, something that can be useful for large animals unlikely to have their feathers preserved in the fossil record. What's more, the detection of feathers associated with Shuvuuia suggests that there may be evidence of feathers associated with some theropods in deposits that would not be considered to exhibit "exceptional preservation." The potential that feathers may surround some dinosaurs in non-lagerstatten deposits requires greater care and attention to detail while excavating. In such situations where the presence of feathers might be ambiguous chemical tests can help resolve the issue, the presence of feathers on Shuvuuia (Schweitzer et al 1999) confirmed when tests turned up beta-keratin. This simultaneously placed Shuvuuia close to birds and illustrated that proteins may survive much longer than expected in the fossil record.

As Mark Norell once explained to me when I asked him about this question we can be as confident that dromeosaurs had feathers as we are that Australopithecus was covered in hair (a statement similar to one attributed to him in Science in 1999). Reconstructions of feathered dinosaurs may sometimes look silly, yes, but our aesthetic preferences should not dictate whether we accept or reject a scientific reality. Dromeosaurs, and possibly most coelurosaurs, had feathers, and that is one of the most exciting notions to come out of paleontology in recent years. There are still plenty of questions about how birds evolved even if their familial relationships have become better understood, though, and I have tried to focus on the presence of feathers here rather than their origin. Interdisciplinary studies, particularly of development (Wagner 2005, Vargas & Fallon 2005), have become more important to understanding the origin of birds and old dichotomies have come into question as research continues (Glen & Bennett 2007). As Richard Prum wrote in a 2002 review, "ornithology is extant dinosaur biology," and the origin & evolution of birds will likely be a hot area of research for years to come.


Appenzeller, T. (1999) "T. rex Was Fierce, Yes, But Feathered, Too." Science, Vol. 285 (5436_, pp. 2052-2053

Chen, P.; Dong, Z.; Zhen, S. (1998) "An exceptionally well-preserved theropod dinosaur from the Yixian Formation of China." Nature, Vol. 391, pp. 147-152

Glen, C.L.; Bennett, M.B. (2007) "Foraging modes of Mesozoic birds and non-avian theropods." Current Biology, Vol. 17 (21), pp. R911-R912

Ji, Q.; Norell, M.A.; Gao, K.; Ji, S.; Ren, D. (2001) "The distribution of integumentary structures in a feathered dinosaur." Nature, Vol. 410, pp. 1084-1088

Norell, M.A.; Xu, X. (2005) "Feathered Dinosaurs." Annual Review of Earth and Planetary Sciences, Vol. 33, pp. 277-99

Prum, R.O. (2002) "Why Ornithologists Should Care About the Theropod Origin of Birds." The Auk, Vol. 119 (1), pp. 1-17

Schweitzer, M.H.; Watt, J.A.; Avci, R.; Knapp, L.; Chiappe, L.; Norell, M.; Marshall, M. (1999) "Beta-keratin specific immunological reactivity in feather-like structures of the Cretaceous Alvarezsaurid, Shuvuuia deserti." Journal of Experimental Zoology, Vol. 285 (2), pp. 146-157

Thomas, A.L.R.; Garner, J.P. (1998) "Are birds dinosaurs?" Trends in Ecology & Evolution, Vol. 13 (4), pp. 129-130

Turner, A.H., Makovicky, P.J., Norell, M.A. (2007) "Feather Quill Knobs in the Dinosaur Velociraptor" Science, Vol. 317, pp. 1721

Unwin, D.M. (1998) "Feathers, filaments, and theropod dinosaurs." Nature, Vol. 391, pp. 119-120

Vargas, A.O.; Fallon, J.F. (2005) "Birds Have Dinosaur Wings: The Molecular Evidence." Journal of Experimental Zoology, Vol. 304B (1), pp. 86 - 90

Wagner, G.P. (2005) "The developmental evolution of avian digit homology: An update." Theory in Biosciences, Vol. 124 (2), pp. 165-183

Xu, X.; Wang, X.; Wu, X. (1999) "A dromaeosaurid dinosaur with a filamentous integument from the Yixian Formation of China." Nature, Vol. 401, pp. 262-266

Xu, X.; Clark, J.M.; Forster, C.A.; Norell, M.A.; Erickson, G.M.; Eberth, D.A.; Jia, C.; Zhao, Q. (2006) "A basal tyrannosauroid dinosaur from the Late Jurassic of China." Nature, Vol. 439, pp. 715-718

Zhou, Z. (2004) "The origin and early evolution of birds: discoveries, disputes, and perspectives from fossil evidence." Naturwissenschaften, Vol. 91, pp. 455-471

More like this

By now most people know that feathers are no longer unique to birds. Thanks mostly to a series of wonderful fossils from the Lower Cretaceous Yixian Formation of Liaoning Province, China, we now know that feathers first appeared in non-avian theropods, and were - later on - inherited by early…
I used to receive random unsolicited emails from an individual who strongly promoted the idea that birds could not not not not be dinosaurs, that the entire dinosaur family tree was screwed up beyond belief, that 'dinosaurs' had evolved from random assorted diverse archosaurs, that cladistics was…
Another book review. I've had a lot of them to do lately. The idea that feathers decorated and insulated the bodies of the small bird-like predatory dinosaurs of the Jurassic and Cretaceous - the coelurosaurs - is no longer a speculation limited to controversial artwork, but the mainstream view […
The sculpted skull of the AMNH Deinonychus mount.For nearly as long as I can remember, artistic depictions of Deinonychus and related dromeosaurs have featured the dinosaur as a pack hunter, often pouncing on a hapless ornithischian like Tenontosaurus (see here, here, here, and here for examples).…

You mentioned feathers becoming less likely with increasing body size. I assume that's because of heat regulation in large animals and the fact that the Cretaceous was hot. Aren't there Tyrannosaurid fossils from far northern latitudes? That climate would not have been cold, but would it have been temperate, with a bit of frost in the winter? I would think even a large theropod would be better off with some covering.

Another question is how early did feathers appear? I'm speculating that before feathers there might have been a hairy or bristly covering in earlier theropods. Might Allosaurus have had a buzz-cut?

jck; I mentioned that I think it becomes more unlikely that we're going to find preserved feathers with increasing body size. Larger coelurosaurs might actually be put at a disadvantage in terms of thermoregulation if they were insulated by feathers but I was thinking more in terms of taphonomy. Many of the exquisitely-preserved feathered dinosaurs come from sediments indicative of ash-falls (or in the case of Archaeopteryx the anoxic floor of a shallow, near-shore environment). Smaller organisms will become buried more easily and quicker, larger organisms being exposed for longer unless something really extraordinary happens (as with the hadrosaur mummies). It seems that the bigger the beast the more difficult for exceptional specimens to be preserved, the circumstances it takes to quickly bury a large dinosaur perhaps not being the kind that would preserve feather coverings (if they had them).

As for Allosaurus I don't think it would have had a buzz-cut because it belonged to the other great group of theropods, the carnosaurs. Feathers seem to be restricted to coelurosaurs so I wouldn't expect non-coelurosarian theropods to have them.

The question of when feathers appeared is an interesting one. Since Archaeopteryx already had pretty advanced plumage by the Late Jurassic I would presume that feathers first arose sometimes during the Jurassic and there was a paper about possible feather impressions left around a Jurassic theropod track from Massachusetts. I have only seen the track cited once so I don't know how important it is (or will prove to be) but it would appear that feathers are a very old thing. It's not like what documentaries led me to believe when I was a kid, dinosaurs rapidly evolving into birds at the end of the Cretaceous just in time to escape extinction; there were already birds that lived among many of the known feathered dinosaurs in the Early Cretaceous. As Peter Dodson pointed out we need to reach back further and try to predict where we might find the oldest feathered dinosaurs and the animals close to when birds split. It might be more difficult but there are still many questions about the origin of birds despite what we know about their relationships.

It's not so much a question of "when did feathers originate" as "how far back does the Coelurosauria go?" Archaeopteryx is known from Late Jurassic rocks, sure, but it's a fairly advanced maniraptor. There are plenty of other coelurosaurs that came before it--ostrich dinosaurs, tyrannosaurs, compsognathids, and "general" forms like Ornitholestes. Clearly, coelurosaurs branched off pretty quickly, but did they diversify as quickly? That's hard to say.


Just dismissing them as "general" forms might be the problem, they likely diversified as quickly as they branched off. Consider the fossil record of teeth from the Middle Jurassic, and some of the Morrison and European coelurosaurs from the Late Jurassic as well.

Well, sure. But when I say "diversified," I wasn't referring to the Coelurus-like animals, but coelurosaurs in general. I mean, they MUST have diversified pretty quickly, as maniraptors were around in the Late Jurassic, which means that all the groups which came before (ornithomimids, tyrannosaurs, etc.) must have already been present.

It's interesting that the group didn't really explode until the Late Jurassic/Early Cretaceous, though, and the more basal coelurosaurs didn't get crazy until the Late Cretaceous. I wonder what changed between the Jurassic and the Cretaceous in Eurasia that allowed coelurosaurs to take over the niches previously filled by carnosaurs?

A post like this is pure gold.

I can't wait for your book. Compelling writing!

(Zach, in answer to your question, it was undoubtedly coelurosaur babes. They wanted feathers? They got feathers.)

"Reconstructions of feathered dinosaurs may sometimes look silly, yes, but our aesthetic preferences should not dictate whether we accept or reject a scientific reality."

Speak for yourself, I happen to think the feathered dinosaurs look quite sharp. Infact I cringe a little bit now whenever I watch Jurassic Park or any of the sequels (though the Tyrannosaurs still give me chills.)

Infact as soon as I 'discovered' (through internet research) that birds were living dinosaurs, I went to my friend's farm and 'rescued' a baby starling to hand raise, just so I could say I owned a pet dinosaur. (Silly I know.)

By Max Paddington (not verified) on 21 Jul 2008 #permalink

One minor point Brian, it's Dilong that has evidence of feathers, not Guanlong. I always get the two mixed up as well.

Oh, and while you don't mention it, it is something worth flagging up - contrary to the Nature paper, Juraventor has *something* that resembles filaments preserved on the tail (Goehlich et al. Archaeopteryx, 24, 1-26). So everyone can stop going on about feather-less compsognathids, or at least wait for the monograph before jumping all over them.

Dave; Thanks for the note about Dilong. Admittedly it slipped my mind (much like the feathered oviraptorsaurs) but I did intend to mention Guanlong. I engaged in a bit of speculation given that it hasn't been found with evidence of feathers, that is true, but I did intentionally bring it up as possibly having them if feathers are a mark of coelurosaurs. I should have remembered Dilong though, so thank you for pointing it out.

If only we had fossils to show us if gorgonopsians were furry or not...

Lloyd et al. (in press) seems to suggest coelurosaurs first arose in the Late Triassic. This is earlier than I would have suspected--but I suspect it's based on the ID of Eshanosaurus as a therizinosauroid, given the incredibly long duration of Oviraptorosauria.

Good post...But still i'm not surely convinced that MOST (as you say) Coelurosaurs and Dromeosaurids had feathers;i don't deny that possibility but we have so little material to elaborate valid theories about...

The Velociraptor quill knobs:well,to tell the truth i'm really skeptical about that discovery...You know,it's strange that something similar haven't been found on any other Dromeosaurid skeleton (for example the famous "Raptor vs Protoceratops" fossil) instead,it suddenly popped out with that lil' ulna...Who knows? Maybe the supposed quill knobs are nothing more than fractures or simple bumps due to the process of fossilization.

I repeat,i don't deny the idea of feathered dinosaurs...Only,i need more reliable proof to start believing this theory ;)

I don't know why you find it unconvincing of most dromaeosaurids having feathers, when fourteen genera throughout Maniraptora are known to have feather impressions (and five others showing feathers through other means, such as skeletal or chemical evidence). I don't normally link to Wikipedia, but this section of the article on "Feathered Dinosaurs" summarizes it fairly well (as did Brian's original article-see paragraph #5 above).

I believed feathered dinosaurs existed long before any were ever found (based on their small size and high metabolic rate I believe all dinosaurs had.) It's nice to have the facts catch up with one's suppositions, and also nice to know that artists like Gregory Paul, who had long depicted small dinosaurs as feathered, and was sneered at for his efforts, can go "na-na-na-na-na!" It's probably safe to assume dinosaurs were similiar to mammals in at least one respect: small and medium-sized ones would have been completely or partially insulated, while large and gigantic ones, like large mammals in tropical environments, would be naked skinned, or nearly so.

By Raymond Minton (not verified) on 01 Mar 2009 #permalink