Feb 11 2013

The Evolution of Feathers

Published by under Evolution
Comments: 8

As a follow up to my post last week on feathered dinosaurs, I received this question over e-mail:

How would a creature get feathers in the first place? I figure there would be some intermediary stages between no feathers and fully feathered, but what would these stages possibly be? No other family of species seems to have anything remotely like feathers. Also, what would be the evolutionary advantage of having feathers be specially since the dinosaurs discovered in China were flightless and (to the best of my knowledge) flying dinosaurs like the pterodactyl were already featherless. Any light you could shed would be appreciated.

The e-mail comes from someone who accepts evolution (not a denier), but is genuinely confused about the above questions. This is an excellent question, one that Darwin himself confronted. This also remains one of the common denialist tactics of the creationists, despite the fact that Darwin gave a very cogent answer in Origin of the Species.

The broader question is – how do complex features evolve when their utility would not come into effect until they were far along the path of evolving to their current form? What use is half an eye or half a wing? This question was rebranded in recent decades as the notion of “irreducible complexity,” but the essential question is the same.

The unstated major premise of the question is that a feature must have evolved directly to its current use. Feathers and wings are currently used for flying and so they evolved directly for that purpose. Evolution, however, does not see that far into the future. Features are evolved for immediate use. They must have current utility. Evolution is also opportunistic and chaotic. Features that evolve for one purpose can be adapted to another.

This process was originally called “preadaptation,” but that term fell out of favor because it suggests some anticipation of the later use, which is misleading. Gould and Vrba proposed the alternate term “exaptation” in 1982. This can refer to a trait that evolved for one purpose and then was coopted for another, or to a trait that arose through genetic drift (without a specific adaptation) and then was coopted.

Let me borrow the now famous example from Michael Behe of the bacterial flagella, which he claimed were “irreducibly complex.” It could not function (as a flagellum – which is logically implied but ignored by Behe) if it were less complex than its current form.  Here is an excellent lecture by Ken Miller explaining how the various parts of the bacterial flagella are all homologous to proteins and structures that serve some other purpose. The base of the flagella is the Type III secretory system that some bacteria use to inject toxins into other cells, for example.

So what about feathers? First, let me address the statement by the e-mailer that “no other family of species seems to have anything remotely like feathers.” This is not true. Feathers are an adaptation of the integument (the skin). Terrestrial vertebrates developed a number of different integumentary adaptations. Reptiles have a variety of scales, while mammals have hair and fur. Take a look at the bristly scales on the Atheris hispida snake.

Birds also have integumentary features other than feathers. The wild turkey is a good example – it has a hairy “beard” and skin outgrowths on the neck. Bird beaks and claws and integumentary adaptations, and birds typically have scales on their feet.

Non-avian dinosaurs, from what evidence we have, also had a variety of integumentary features. So – while I agree that feathers are perhaps the most dramatic such adaptation of terrestrial vertebrate integument, they are by no means unique.

The exact evolutionary origin of feathers has not yet been worked out, but scientists are making good progress. The question is – how far back in the history of dinosaurs, or perhaps even pre-dinosaurs, do feathers or their precursors go? This will probably be worked out through genetic analysis.

What purpose could early feathers have served before they were adequately adapted to flight? There are many hypotheses. Since we have no living examples of feathered dinosaurs or similar creatures it will be difficult to definitively demonstrate what purpose early feathers served. What we can do is to come up with plausible options and see that at least they are consistent with the fossil evidence.

One hypothesis is that early downy feathers were an adaption for thermoregulation. Down is very insulating, and would have been very useful for keeping warm, especially for young dinosaurs. These feathers could then have been coopted and adapted for mating displays. Broad feathers could also be used for trapping insects for food. More directly related to flight, early feathers could have been used to increase the range of predatory pounces, and even provide for some mid-air correction. They could also have been used to slow and guide the descent when dropping from a height, such as a tree limb. This could have led to gliding flight, and finally to full flapping flight.

The above examples are not a literal linear sequence, and also adaptive radiation means that feathers may have taken many different paths through various adaptations and uses, at least one (probably more) eventually leading to full flapping flight.

This at least answers the question – how could feathers have evolved before they were developed enough to be useful in flight. It turns out that half a wing could have had many uses.

8 responses so far

8 thoughts on “The Evolution of Feathers”

  1. juga says:

    A related question. Are all present day flightless birds thought to have evolved from flying birds? If not, studying the purpose of feathers in flightless birds might help understand their origin.

  2. Yes – all flightless birds are descended from the common ancestor of birds that was flying. They are therefore not a perfect analogy to paraves species that could never fly.

  3. daedalus2u says:

    Porcupines have quills that are quite feather-like in that they have a hollow shaft.

    Some horns are keratin and hence from the skin (cows and rhinos) rather than bone as in antlers.

  4. AndrewTyson says:

    Do we have any examples of early feathers? The irreduceable complexity argument was pretty soundly countered with the example of the eye starting from a patch of light sensitive skin and slowly changing shape, each step of the way becoming more useful. While we might not know the use for an adaption of an early feather, do we know of an example of them existing. Would they have evolved from what we think of as a quill? How similar in structure is it?

  5. ZooPraxis says:

    I really enjoy these pedagogical posts and would even like tips on pronunciation.
    I don’t know if they bore or frustrate your more educated readers but as a layman with no background in science I find them really useful.

  6. lfvvb says:

    Indeed, the randomness and historical baggage is evidence of evolution. There is also the sexual selection hypothesis, since the paleobiologists could infere some of the fossil feathers colors.

  7. ccbowers says:

    “Yes – all flightless birds are descended from the common ancestor of birds that was flying. They are therefore not a perfect analogy to paraves species that could never fly.”

    I wouldn’t necessarily view this as a flaw in the analogy for this discussion, because this fact demonstrates a major flaw in the irriducibly complex argument. Not only is it often an argument from incredulity, but this incredulity exists due to an assumption that the utility of the feather (or eye or whatever) remains constant over time.

    The fact that penguins do not fly, but have feathers that are vital for temperature regulation, yet these feathers come from an ancestor that used them to fly (and also perhaps temperature regulation), which in turn had ancestors that had protofeathers for other purposes other than flying really brings the point home that the concept of exaptation is essential to understanding these questions.

  8. a_haworthroberts says:


    You may have seen a recent anti-evolutionist blog post by Dr Jay Wile:

    Some of what he says seem fair enough (concerning a paper on Sinosauropteryx dating from 2012) but he also writes: “At university, I was taught as fact that feathers evolved from scales. However, we now know that there are several reasons why that isn’t possible”. He does not elaborate on this sweeping claim, and simply links to this article from 2003:

    However – without subscribing – I cannot read this beyond the first two paragraphs. Perhaps you have read it? Of course reptile scales, like feathers, are made of keratin.

    I’ve also just read this 2011 National Geographic article (I expect you are familiar with it):

    The article does not state that feathers evolving from scales isn’t possible. It does say:
    “The long, hollow filaments on theropods posed a puzzle. If they were early feathers, how had they evolved from flat scales? Fortunately, there are theropods with threadlike feathers alive today: baby birds. All the feathers on a developing chick begin as bristles rising up from its skin; only later do they split open into more complex shapes. In the bird embryo these bristles erupt from tiny patches of skin cells called placodes. A ring of fast-growing cells on the top of the placode builds a cylindrical wall that becomes a bristle.
    Reptiles have placodes too. But in a reptile embryo each placode switches on genes that cause only the skin cells on the back edge of the placode to grow, eventually forming scales.”

    It seems that creationists tend only to quote the most up-to-date papers or articles when their contents appear to be helpful to them.

    Ashley Haworth-Roberts

Leave a Reply