The Other Side of the Raptor?

Velociraptor! Utahraptor! Deinonychus! Among the more popularly illustrated dinosaurs, perhaps none other are subject to such a wide variety of artistic interpretation, especially concerning their integumentary structures. Last year I posted an essay on this site to explain why arguments against the feathering of Dromaeosauridae generally fail [1]. Since then, there have been a few more additions to the public presentation of dromaeosaurs, but old myths persist. In the teaser preview for Peter Jackson's 2005 remake of King Kong, we see carnivorous dinosaurs that look suspiciously like the bare-skinned raptors of Jurassic Park and its imitators [2]. (I have not followed the production of this movie very closely, but I read that the dinosaurs are meant to be fictional species evolved in the 65 million years since the Mesozoic.) Looking back on my previous essay, I believe my review of the conflict makes a valid point. But because I feel like writing today, allow me to present a somewhat different view...

When we talk about the "right" way to restore the integument of dinosaurs like Deinonychus, we tend to be speaking in relative, not absolute, terms. To understand why, we need to look at the philosophical foundation that underlies many feathered dinosaur illustrations, the inference of feathers from related forms. The preferred multiple-outgroup approach, called the Extant Phylogenetic Bracket [EPB] method by Witmer, is explained fully in articles by Harold Bryant and Anthony Russell [3] and by Lawrence Witmer [4]. While the approach is undeniably logical and useful, it is not meant to be confused with omniscience. The EPB method can at best only provide us with the most likely solution, not a full range of possibilities and impossibilities, and not the guaranteed Truth. Before we dismiss an illustration of a bare-skinned dromaeosaurid as "wrong", we should be aware if our concept of "wrong" in this case means "impossible" or "somewhat less likely than another option", and adjust our reaction accordingly. It is quite a different kind of "wrong" from an illustration that is flawed in some known skeletal detail, such as a Deinonychus improperly restored with a five-fingered hand.

In my previous essay, I asked for suggestions as to why natural selection might favour a featherless Deinonychus. These things would be nice to know, but cannot be expected in palaeontology. Adaptive scenarios- the story of what a particular trait evolved "for"- have been criticized as untestable assumptions that often reflect human prejudice and even hinder our search for true evolutionary patterns. The argument against adaptive scenarios has been discussed in depth by Henry Gee [5]. A cladistic view of dinosaur evolution is produced by a search for shared traits, not by telling stories about why these traits occurred. The evolution of dinosaurs was not constrained by human creativity or understanding, so some evolutionary changes that likely occurred may not be easily written into an appealing story. If a bare-skinned dromaeosaurid is ever discovered, we will probably speculate on how it arrived at this unusual condition, but the story will not be essential in considering where the new specimen fits in dromaeosaur evolution.

When it comes to featherless coelurosaurs, scientists and artists do make a few exceptions. Tyrannosaurus rex has sometimes been imagined as a secondarily featherless dinosaur, especially in light of its feathered tyrannosauroid relative Dilong [6]. If true, this would be good news for nearly all existing paleoart, which depicts T. rex without feathers. Not much is known about T. rex integument [7], but the idea of secondary feather loss is always justified by the adaptive scenario of T. rex being so large that it would have overheated had it been feathered. But is this a hypothesis based in science? I am not aware of any technical mathematical study predicting the theoretical maximum size for feathered animals [8], and such a study would face many uncertainties regarding the thermoregulatory strategies of tyrannosaurs. And even if T. rex does turn out to be secondarily featherless, we may have the scenario wrong- we cannot yet be certain that tyrannosauroids did not lose their feathers before evolving large size, however less intuitive this seems.

The real law being preached is that relatively small coelurosaurs must maintain feathers. We can never be completely certain of this statement because we are unlikely to examine the integument of every small coelurosaur to ever exist. Actually, we cannot examine the integument of most small coelurosaurs- most of our experience is limited to extant Neornithes. It would be odd to believe that living birds encompass the full range of possibilities for dinosaur evolution. There is no shortage of examples of extinct dinosaurs evolving unique features not paralleled in birds, and perhaps one day a small, secondarily featherless coelurosaur will join the list. When we hear that dromaeosaurids absolutely cannot be featherless, this is something that we have told the dromaeosaurids, not something that the dromaeosaurids are capable of telling us.

Belief that featherless dromaeosaurids are outright impossible may really be missing the point of cladistic philosophy. Feathers are a major contribution to the superficial appearance of an animal. When we see a feathers on a dromaeosaur, this is a powerful instant reminder of the theropod's close affinity to birds. But we may be unfortunately close to regarding feathers as a "key" character of birdlike dinosaurs, one immune to the normal processes of evolution and reflecting some perfect archetype of what a bird or dromaeosaur should be [9]. The typological mode of thinking conflicts with the current recognition that we cannot know which, if any, characters are more important than the rest, so all must be considered to avoid unscientific assumptions. Therefore, we see phylogenetic studies in which a character, "Feathers: absent (0); present (1)", is considered no more influential- and no less prone to reversal- than less emotionally-invested characters involving tooth serrations or the relative lengths of wing bones [10]. If we find that the presence of feathers did reverse to absence at some point in dromaeosaurid evolution, it would indeed make for a very surprising discovery. However, we must also remember that it would not be fundamentally different from any of the less exciting reversals already discovered in dinosaurs.

What do the fossils tell us? In a sense, nothing, because no unequivocally featherless domaeosaur has been found. When I recieved the new cladistic analysis of dromaeosaurs by Senter et al. [11], I noticed that more than a few of the diagnostic traits within Dromaeosauridae are reversals of the traits diagnosing more inclusive coelurosaur clades. While I am aware that the aim of cladistics is to minimize the total number of reversals, I cannot help but wonder if all of the reversals involved in making a dromaeosaurine were restricted to the skeleton, or if there were also substantial reversals in other areas such as feathers. We simply do not know. What we do know about dromaeosaurs is that while all integument-informative specimens have feathers, the precise pattern of feathers seems to vary among taxa. Microraptor has the most elaborate plumage, but Sinornithosaurus may lack flight feathers. It is unknown if either of these animals represent true extremes in the range of variation for dromaeosaur feather distribution.

If I had to choose just one image of Deinonychus as my best guess of its true appearance, the feathered version is a much safer bet. Featherless dromaeosaurs probably did not exist- but if anyone chooses to restore them that way, a bit of unsupported speculation is not a crime. Dromaeosaurs without feathers are not violating any rules of how dinosaurs may have evolved; nor do they necessarily imply a more distant relationship to birds. Dinosaur palaeontology has a history of shocking and ironic twists. Do not be too harsh in mocking the artists who still create featherless dromaeosaurs, they may have the last laugh.

Notes and References

1. "Dromaeosaurs with Feathers: Some, or All?"

2. King Kong trailers

3. Bryant, H.N. & A.P. Russell, 1992. The role of phylogenetic analysis in the inference of unpreserved attributes of extinct taxa. Philosophical Transactions of the Royal Society of London B337: 405-418.

4. Witmer, L.M., 1995. The Extant Phylogenetic Bracket and the importance of reconstructing soft tissues in fossils. 19-33 in Thomason, J.J. (ed.). Functional Morphology in Vertebrate Paleontology. Cambridge University Press, Cambridge.

5. Gee, H., 1999. In Search of Deep Time: Beyond the Fossil Record to a New History of Life. Cornell University Press, Ithaca and London.

6. Norell, M.A. & Xu X., 2005. The varieties of tyrannosaurs. Natural History 114 (4): 34-39.

7. Small patches of unfeathered skin impressions are said to exist from tyrannosaurids, but detailed descriptions have not yet been published.

8. Does a paper like this exist? If you know of it, I would very much appreciate it if you told me the citation.

9. A less strictly applied "key character" of Aves is flight, which we accept as reversible. But if flightless birds were for some reason all extinct, and had not yet been discovered as fossils, one can imagine the concept of a "flightless bird" being as uncomfortable as the concept of a featherless bird.

10. For example, see Chiappe, L.M., 2002. Basal bird phylogeny: Problems and solutions. 448-472 in Chiappe, L.M. & L.M. Witmer (eds.). Mesozoic Birds: Above the Heads of Dinosaurs. University of California Press, Berkeley and Los Angeles. Feathers are character #168; for the taxa chosen, this character is uninformative.

11. Senter, P., R. Barsbold, B.B. Britt & D.A. Burnham, 2004. Systematics and evolution of Dromaeosauridae (Dinosauria, Theropoda). Bulletin of Gunma Museum of Natural History 8: 1-20. This is one of several current competing theories of dromaeosaur relationships.

Back