A Blog by Ed Yong

Dramatic restructuring of dinosaur feathers revealed by two youngsters of same species

SimilicaudipterxAt the Chinese Academy of Sciences, Xing Xu is looking at two beautiful dinosaur fossils, both with clear feathers on their arms and tails. In the smaller specimen, the feathers are like thin ribbons at their base and quills at their tips (with vanes coming off a central shaft). The larger specimen is different – its arm and tail feathers are like quills across their entire length.

With such different feather structures, you might assume that these animals belonged to different species, but you’d be wrong. They’re actually different life stages of the same animal – Similicaudipteryx. Both are youngsters, but the one with the quill-like feathers is an older version of the one with the ribbons. Together, they demonstrate that the feather of some dinosaurs changed dramatically as they grew older, in a way that we don’t see in any modern bird.

By now, readers of this blog should be familiar with the idea of feathered dinosaurs (and, indeed, Xing Xu has discovered many of them). A spectacular series of fossils have revealed a wide range of plumes in a wide range of species, and we even know something about their colour. But we still know very little about how these feathers developed as the animals matured, because fossils of young feathered dinosaurs are few and far between. So for Xu to find two, and two of the same species no less, is a real treat.

Both hailed from Liaoning province of China (where else?), and based on their skulls, spines and hips, Xu has confidently classified them both as Similicaudipteryx, a small predator from the oviraptosaur group. Both animals are clearly youngsters. Although one is larger than the other, they’re both smaller than adult specimens of the same dinosaurs, and some of their bones haven’t fused completely yet.

The younger animal (a-c below) has downy feathers over much of its back and hips. Elsewhere, it has larger pennaceous feathers (with a shaft and vanes) – 10 on each arm, and 11 much larger ones on its tail. All of these are ribbon-like at the base and quill-like at the tips. The more senior juvenile (d-f below) also had downy feathers on its head, back and hips but its pennaceous feathers are very different to its younger peer. Each arm has 10 primary feathers and 12 secondary ones, and the tail had at least 12 pairs. All of them are quill-like from base to tip and the arm feathers are just as long as the tail ones.

Similicaudipterx_youngSimilicaudipterx_olderThese differences suggest that Similicaudipterx’s feathers changed dramatically as it grew from hatchling to adult. The move from partial ribbons to complete quills is the most obvious one, especially since the arm and tail feathers of modern birds hardly change after they moult their initial downy birthday suits. It’s possible, but very unlikely, that the individual feathers changed; instead, Xu says that the animal probably moulted its feathers as it grew and replaced them with new ones of a different type.

There were other changes too. If the younger animal really lacked secondary arm feathers (and it’s possible these just didn’t fossilise well), then Similicaudipteryx must have grown these as it matured. Modern birds have them from hatching. And the fact that the tail feathers outsized the arm ones in the younger individual but not the older one suggests that the feathers developed at different rates. Xu suggests that the arm feathers became more important as adulthood loomed.

We can probably even guess the genetic events that lay behind these changes. In modern feathers, scientists are well aware of the genes that control the formation of the rachis (the central shaft) and the barbs (the branches that come off the rachis). If you switched off genes that promote the production of barbs, such as sonic hedgehog (don’t ask – developmental biologists make their own fun), you’d get a flat continuous ribbon on either side of the rachis. That’s probably what happened in the younger animal, resulting in a feather that’s half-ribbon and half-quill. These genes were then switched on more forcefully in the older juvenile and in adults.

The half-ribbon feathers, formally known as “proximally ribbon-like pennaceous feathers” or PRPFs, are found in other dinosaurs, including primitive birds like the confuciusornithids. But they’re nowhere to be seen in modern birds. Because of this, Xu says that the development of the earliest feathers was probably much more varied and flexible than in today’s birds. It takes the discovery of rare and stunning specimens to reveal these programmes, which have since been lost in the course of evolution.

Reference: Nature http://dx.doi.org/10.1038/nature08965

More on feathered dinosaurs:


Twitter.jpg Facebook.jpg Feed.jpg Book.jpg

13 thoughts on “Dramatic restructuring of dinosaur feathers revealed by two youngsters of same species

  1. Forgive my ignorance, but I don’t understand the idea that “the feather of some dinosaurs changed dramatically as they grew older, in a way that we don’t see in any modern bird.” Don’t hatchlings have downy feathers until they are fully-fledged adult birds? Please do explain.

  2. This critter looks a lot like Incizivosaurus (=Protarchaeopteryx)

    Great article, Ed. Can’t wait to read the paper. Those are some beautiful fossils.

  3. There’s a skull of Caudipteryx (currently referred to Caudipteryx zoui) that looks superificially like that shown here, but without the buckteeth. The buckteeth appear to be highly derived. It may also help disprove Senter’s theory that Incisivosaurus is a skull type that belongs to a Protarchaeopteryx-like animal, or help reinforce it by putting Protarcheopteryx at the base of Caudipteridae. Fun stuff.

  4. @ Amy Heller
    Dinosaurs: ribbons –> quill-like
    Modern birds: downy feathers –> fully-fledged
    Both change, but not in the same way, hence “a way we don’t see in any modern bird”

  5. And bear in mind that the young Similicaudipteryx isn’t exactly a “chick” so it might well have gone through a downy stage. In modern birds, after the initial moult that gets rid of the downy birth-feathers, the animal has the same structure of feathers for the rest of its life. These fossils suggest that feather structure continued to change in dinosaurs, making at least one further transition from ribbons to quills.

  6. Would love to see the chart with feather morphologies and dino phylogeny that Nature teases us with in the abstract, but Nature purposely has it at such a resolution that one can’t read it. Oh well, I guess that’s why God created libraries — for us impoverished types who can’t afford all the subscriptions required.
    BTW, slightly off topic, but related to your earlier guest column on science reporting and getting both sides, I was slightly taken aback by a reaction published in the Nature News roll out of the piece. It quotes “developmental biologist” Cheng-Ming Chuong from the University of Southern California in Los Angeles, who begins the assessment with “If we give [the?] authors the benefit of doubt …” Perhaps this was a poor choice of words, but it seems to me that it falls in the category of “unnecessarily confrontational.” Wouldn’t it have been better to say “based on what I’ve seen in the paper”? After all if Xing Xu isn’t entitled to the “benefit of the doubt” on feathers, then I guess science discussions on the net have descended to the level of political discussions on the net.

  7. This is peripheral to the development question, but is related to the feather discussion.

    The more senior specimen has arm feathers that really are similar to the superficial aspects of current wings. The pictures indicate that these were agile runners and they certainly have claws. How do the wing-like feathers fit in terms of function, whether in capturing prey or in some other way?

  8. such as sonic hedgehog (don’t ask – developmental biologists make their own fun)

    Ah, but the story behind the names is sometimes part of the fun for others. E.g. the gene named “indy” which when mutated doubles the lifespan of the fruit fly. The name is an acronym for I’m Not Dead Yet–referring to a running joke from Monty Python and the Holy Grail…and if you haven’t seen that film, then you simply must. How can you understand what goes on in science if you haven’t almost memorized the dialogue from that film? 😉

    Interesting how the primary feathers number 10, same number as most passerines, and many non-passerines.

  9. While I agree that the evidence could suggest that this dinosaur changed their feathers over its lifetime, isn’t it shortsighted to jump to that conclusion. The article mentions that finding these juvenile fossils is incredibly rare, so how can we speak for the species based on two individuals.

    It is equally likely that these two types of feathers are simply two difference phenotypes (physical traits) of the same species. Genetic variations can be found in this population, so the younger dinosaur could have inherited a different feather gene than the older dinosaur. It doesn’t necessarily mean that an individual’s feathers change over time.

  10. Dan, given the analogy to modern birds going through several moults, I think the more reasonable answer for now is that the younger animal really is a younger animal at a different plumage stage. It would be telling if somebody looked at the feathers of “Dave” (usually considered a juvenile Sinornithosaurus) and then compared it to the feathers of adult fossils.

    To answer Mike’s question, the functional aspect of arm-feathers in non-avian dinosaurs has always been up for debate. Prey capture probably has little to do with it. You may be referring to Ostrom’s old idea that Archaeopteryx used its wings as “bug-catching” scoops. This most certainly did NOT occur. It’s more likely that large arm feathers actually hindered prey capture. I would expect that large predatory maniraptors (like dromaeosaurines) had reduced manual and arm feathers. Some of the more interesting “reasons” for arm feathers that I’ve read have been parachuting or flying (ancestrally), brooding, and sexual display. I can’t imagine why all three of these functions couldn’t have overlapped, either.

    Notice, also, that pennaceous feathers are unique to “higher” maniraptors. Alvarezsaurs and therizinosaurs didn’t have pennaceous feathers, at least not on their arms. I don’t think there’s good evidence that oviraptorosaurs were carnivorous, so maybe their long hand and arm feathers didn’t get in the way when feeding.

  11. Amazing stuff coming out of China these days. They seem to be leaders in the area of feathered dinosaurs (both discovery and analysis).

  12. About the “half-ribbon feathers”.
    During moult, long feathers have a coating to bundle them up, and the vanes are spreading out from the top down. Are the “half-ribbon feathers” not simply pennaceous feathers caught in their first growth during moulting?

Leave a Reply

Your email address will not be published. Required fields are marked *