Monday, February 14, 2011

Heat, Feathers, and Half-Arsed Velociraptor

Above: Comparison of common Velociraptor reconstructions by artist Tomozaurus, used with permission.

A month or so ago, this diagram (also shown above) produced by an artist known as Tomozaurus stirred up a good deal of debate over at DinoForum. Tom was trying to illustrate common misconceptions about the most likely life appearance of dromaeosaurids. By now, everyone (including the birds-are-not-dinosaurs crowd) agree that dromaeosaurids were fully feathered and pretty bird-like. Whether or not to consider them actual birds is a matter of semantics at this point. But, among dino-fans weaned on their depictions in pop-culture, there seem to be a lot of heated reactions to depicting them as too bird-like. Many will admit that they had feathers, but stop short at reconstructing them in a really bird-like manner, preferring a short, cat-like pelt that allows the graceful and well-known contours of the skeleton to show through in life. But compare any bird skeleton to a live specimen, even those with "simple" feathers like chicks or kiwi, and you'll immediately see that this is the wrong way to go. Even most feathered theropod fossils show a feather covering that does not hug the body contours, but like modern birds, consists of a lot of poofy, long feathers (especially at the breast and neck) that would make them look very "bulky."

The debate about the above image concerned how well these principles should apply to larger dino-birds like Velociraptor. Velociraptor is perhaps the worst offender in this area due to its enormous popularity: it has a very well-ingrained image in the popular consciousness that, in all likelihood, doesn't match how it would have appeared in life. But just how likely is any reconstruction?

The fact is, we don't know very much at all about the feathering in Velociraptor. At least one specimen preserves quill knobs on the ulnae where large secondary feathers must have attached, so we know for a fact that it had wings. But what about the body feathers? Tom's picture supposes that we would do best to reconstruct the remaining feathers based on related (but much smaller) species in which the full compliment of feathers has been preserved; things like Microraptor, Anchironis, Archaeopteryx, and the unnamed species nicknamed "Dave". Based on these, Tom gave his "correct" Velociraptor a full set of long primary feathers (present in Micro and Archie but much shorter in Dave and Anchi), a feather crown (present in Micro and Anchi but not Dave and Archie), flight feathers on the hind legs (present in Micro, Anchi and Dave but not Archie, though the last does have "bell bottoms" of long, non-planar pennaceous feathers down to the ankles), and feathers covering the face and most of the snout (present in all but possibly Archie). Last, he restored the body contour feathers as pennaceous rather than downy. This is the most problematic, present (apparently) in Anchiornis but not in the others (Archaeopteryx has pennaceous feathers over the hips but this appears to be a single or paired feather tracts (pterylae), and the rest of the body is covered in plumulaceous feathers or simple protofeathers).

In contrast, Tom presents two "incorrect" versions: the Greyhound/Lizard (which is so obviously wrong there's not much else to say about it) and the "Half-Arse." The later is interesting because while it deviates considerably from the inferences drawn based on its relatives, there's nothing obviously inaccurate about it, or at least implausible. The drawing seems to be identical to the Velociraptor computer models used in the TV show Dinosaur Planet: feathered, but barely so, with a short, mammal-like pelt hugging the contours of the body, a large crown for display, and many featherless areas of the body.

Defenders (or at least devils advocates) for this reconstruction pointed out that it doesn't deviate too wildly from larger, flightless modern birds. Ostriches (Struthio camelus) are famously stripped-down of feathers, more so than many people think (see image above, showing the naked underside of the wings and featherless torso). However, ostirches are also quite a bit larger than Velociraptor, and smaller ground birds from the same environment like the Secretary Bird (Sagittarius sepentarius) lack extensive naked patches. On the other hand, Rhea (Rhea spp.), which are similar to Velociraptor in weight, do have naked patches on their underwings, though not as extensive as ostirches (see image below).

So, while I still think the rather short body feathers would be a little far-fetched (longer ones would be better for thermoregulation--if overheating is the issue, the feathers would probably simply be lost entirely, which is not the case in modern birds that lose feathers only on extremities), the Half-Arse version doesn't seem all that bad.

But there's more that needs to be taken into account. A great post at Tetrapod Zoology today touches on some aspects of featherless patches on birds and its effects on thermoregulation. A bit counter-intuitively, a heavy feather coat can actually help keep birds cool--they're general insulators, not simply heat-trappers, like a thermos. In the post, Darren Naish discusses male wild turkeys, which famously have bare heads with brightly colored, outlandish soft tissue structures for display. This actually puts the male birds at a disadvantage, because all that bare skin causes the males to be more prone to overheating, requiring them to spend more time cooling in the shade than females. Like the tail of a peafowl, the naked heads of turkeys are a sexual display structure that puts the birds at a disadvantage when it comes to survival. Granted, this is only the example of one bird. I don't know how this might apply to, say, ostriches. Ostriches, like Velociraptor, inhabit a hot, arid environment. Indications suggest that Velociraptor lived in an even more desert-like setting, dominated by barren dune fields, making it a solidly desert bird. Would the extensive bare patches of the Half-Arse be beneficial in this setting, or would they tend to drastically overheat the animal under an unrelenting desert sun with few sources of shade?

Above: Photo of a Greater Rhea (Rhea americana) by Ramon Moller Jansen. This species is about the same size as an adult Velociraptor mongoliensis, but live in more lush environments with some tall shade plants.

According to Willmer, Stone & Johnson (2000), many ratites use similar strategies to regulate their body temperature and prevent overheating. Ratites such as the Rhea store their heat while active, and only actively attempt to shed it while at rest. This is achieved through strategies such as panting, drooping the wings (allowing air to conduct heat from the sparsely feathered underwings while at the same time shading them), and raising or lowering the sparse feathers of the back (known as ptilo-erection). It's interesting to note that in Archaeopteryx, the back is the only region of the body that has pennaceous feathers, better for trapping or shedding heat. With all of these adaptations, ostriches rarely have to seek shade, even when it's available (Levy et al., 1990). So while feather-loss in modern birds is an important strategy for those living in hot, arid climates, note that the pattern of loss is not random or extensive, but rather strategic, allowing for maximum thermal regulation.

If I were to speculate, I'd make an educated guess that bare patches in Velociraptor must have been limited to the legs, flanks and underwings for these reasons. Those areas could be shaded simply by the animals own body and cooled easily as needed. The head and neck would have been most prone to overheating. The animal could only shade these by adapting a Mei long style sheltered posture with the head tucked under the wing and body feathers (but only if a hefty body covering was present, not a form-fitting pelt). One question I can't seem to find an answer to is how ostriches cope with direct sun to their nearly bare heads and necks--the problem faced by male turkeys. My wild guess would be that the heads of turkeys are highly decorated with thickened skin, presumably a better insulator than the plain skin of an ostrich, which would allow more loss of heat to the air. Or maybe they bury their heads in the sand to keep cool... (I kid, I kid!). If anybody knows of any studies or physiological issues that would cause this to be a problem for one and not the other, please leave a comment!

* Levy, A., Perelman, B., Grevenbroek, M.V., Creveld, C.V., Agbaria, R. and Yagil, R. (1990). "Effect of water restriction on renal function in ostriches (Struthio camelus)." Avian Pathology, 19: 385-393.
* Willmer, P., Stone, G. and Johnston, I.A. (2000). Environmental Physiology of Animals. Wiley-Blackwell, Science. 644 pp.


  1. This blog just took a level in awesome. If that were possible.

  2. "...there's nothing obviously inaccurate about it, or at least implausible."

    Note the pronated hands and lack of primary feathers. I'd also claim the naked pubic boot and underside of the tail as fairly implausible.

    Nice post. It gives some additional helpful information I can include on my second version of the image.

  3. There's also behavior to consider. Covering the muzzle in feathers, even partway down like the artist has in the "real" version, may not have worked for a predatory animal that's going to have to be plunging its face into a kill. An extreme example would be today's vultures, though I doubt a Velociraptor would have had the fully bald head and neck.

    It's been long suggested that Velociraptor's retractable claws indicate that it took relatively large prey in comparison to its body size. That could potentially suggest the behavior of probing into the prey's body cavity for nutritious viscera, an thus the need to keep portions of the face free of feathers.

    Just a thought.

  4. @Tabbyraptor:
    It's certainly possible, although again, it's important to consider the function behind such adaptations. Pennaceous feathers are more difficult to clean for modern birds, especially on the head where preening with the bill is impossible. However, the head and most body feathers of smaller dromaeosaurs were not generally pennaceous (except for the crown in some species), rather they were filamentous. Those structures would be more analogous to mammalian hair, and in mammals you don't really see fur loss due to carnivory. In fact not even all vultures have bald heads, and I've read (can't think of where) that the typical bald head of turkey vultures and condors may be as much for display as for hygiene.

  5. @Matt:
    That's an excellent point. I would give a lot to know whether the animal had a thicker, flexible mammal-like tongue, or a thin, relatively immobile bird-like one. That would make a fairly critical difference in hygiene practices.

  6. @Tabbyraptor
    I would expect inflexible considering that all living theropods (that I know of) share this feature.

    1. I'm not sure that's true, considering the huge amount of muscular control in the tongues of creatures like psittacenes and finch-like birds? I don't know much about the level of flexibility, but the size of bird tongue-to-beak is hugely variable (kookaburras have a ridiculous little nub, while parrot tongues fill the whole beak). I have also observed that pigeons have rather large tongues and wet mouths (they even use it as a communication--very content, settled pigeons do a terribly cute thing of sticking the tip of the tongue out from the end of the beak and in again repeatedly, which is reminiscent of the beakgrinding famous in contented parrots). There is a great deal of variation in avian tongues, either way.

  7. I wonder if protofeathers could aid in gliding and/or flight. I'm just thinking of that because a lot of scientists say microraptor could glide. Microraptor is closely related to velociraptor I would assume since they're both raptors. If raptors only had protofeathers, could they glide with that? I wonder what type of feather microraptor had. Is it a more evolved type of feather than velociraptor, I wonder? I doubt anyone here knows the answer, but I felt compelled to muse publicly.

    By the way, I recently started blogging about dinosaurs myself and I'm following your blog. I guess you could call me another "armchair paleontologist". I checked out your website and it's pretty cool. I'm actually working on a dinosaur website myself right now for my web design class. When its done, would you like to take a look at it? It's supposed to be about how to draw dinosaurs. I'm doing my best to make it accurate if I can by doing research on dinosaurs online. I'm not a professional paleontologist, obviously, so I'd like as many bits of help and suggestions I can get on my research. Do you mind if I e-mail you so you can take a look at my research so far?

  8. @Holly Amber
    Actually, neither Microraptor nor Velociraptor are known to have "protofeathers." Microraptor has feathers identical to those o modern birds, and by all indications, so did Velociraptor. Certainly, modern bird feathers are conducive to flight or at least gliding. All animals we know of that may have had "protofeathers", the feathers (really more like modern chick down than "proto"-anything) are either very short (again, think of the wings of a baby chicken) or long and quill-like (like a porcupine), so would not have been much use in flight, though there may be intermediate forms we haven't found yet with long, downy wing feathers that would lend some aerial ability.

    Look forward to seeing your blog!

  9. Magnificiently well written, and most informative. Thank you.