New Bird-Like Dinosaur Settles Origins of Flight

By admin | May 29, 2013 12:11 pm

By Jon Tennant

Reconstruction of Aurornis xui, a new basal avialan from the Middle/Late Jurassic of China. Credit: Masato Hattori

Feathered dinosaurs used to be as valuable as gold dust. Now, so many specimens have been unearthed that museums are overflowing. But for all the specimens, a crucial question has remained unanswered: which species was the original ancestor of birds?

A new species found in China has shed light on the answer. The two-foot long Aurornis xui, the “daybreak bird,” fleshes out the relationships between bird-like dinosaurs and, along with its cousin species Archaeopteryx and Anchiornis, restores its lineage as the likely predecessors of birds.

The early relationships of a group called Avialae, the dinosaur line leading directly to modern birds, have been a hot point of debate. Over the last few years, with new bird-like dinosaur discoveries, some palaeontologists have even removed the iconic Archaeopteryx and its relatives from the Avialae group altogether. However, this shift would have meant that powered flight evolved multiple times in feathered dinosaurs, a less likely situation than one such adaptation.

Skeleton of the new species Aurornis xui. Credit: Thierry Hubin/IRSNB

Now, Aurornis xui, named in honor of dinosaur hunter Xu Xing, appears to have settled that part of the debate. Along with Anchiornis of the same age (about 155-160 million years), and Archaeopteryx (150 million years), Aurornis fits within the earliest ancestry of Avialae. The existence of three contemporary species allowed researchers to triangulate their relationships based on their appearance. Archaeopteryx was knocked off its perch as the oldest bird back in 2009 by Anchiornis, and now Aurornis has finished the dethroning, with the three species together sitting at the very base of the Avialae lineage that would later give rise to all other birds.

The new research also shifts the position of Balaur, a flightless Romanian dinosaur, from the non-bird line dromaeosaurids (the group including Velociraptor) to a much firmer placement within Avialae. Aurornis, like Balaur, lacks pennaceous feathers. Both were thus incapable of flight, unlike the rest of the Avialae.

This positioning of Balaur, Aurornis and Archaeopteryx within Avialae means that powered flight only evolved once in dinosaurs. Feathers thus almost certainly initially evolved for things other than flight, such as for signalling, or keeping eggs warm when brooding.

Still, it’s unlikely that this is the final shakeup in birds’ early history, says study co-author Gareth Dyke from the National Oceanographic Centre in Southampton, UK: “As more fossils are found in this unexplored age, these relationships are going to change – it’s currently in a great amount of flux.”

MORE ABOUT: archaeology, dinosaurs
  • Inane Rambler

    It’s unlikely we’ll actually find the species that lead to birds, but we’ll probably find a lot of it’s relatives.

  • David Bump

    We could find all sorts of feathered dinosaurs and feathered, flying dinosaurs, but that still wouldn’t prove that a line of dinosaurs could or did evolve into birds. Still, if you have faith that such an amazing transformation could and did take place, this certainly must be encouraging.

    On the other hand, how many beautiful, whole-body but singular specimens in this series are going to turn up in an area the size of one state, without any scientist ever seeing one in the ground, and hardly anything like them turning up anywhere else in the world, before it becomes reasonable to wonder if somebody there has come up with a fool-proof way to make fake fossils to order? Remember, “Archaeoraptor” got as far as National Geographic publication before it was discovered to be a rather clumsy hoax with an added-on tail…

    • Smartass

      A fossil-hunter made a hoax and National Geographic (which is not a peer-reviewed journal) published it. Scientists exposed the deception. Archaeoraptor isn’t a reason to mistrust science.

      As for “proof” of dinosaurs evolving into birds: you won’t find it and never could. They might not have done, but then again you might be asleep and dreaming that you are reading this, and your whole life might be a dream, and the rest of the world might not really exist. You can’t “prove” otherwise.

      But for practical purposes, what matters is what theory best explains the evidence collected to date. Evolution is the clear winner so it is reasonable to consider it true.

      • David Bump

        Sorry I’m so late, been busy IRL and had major problems with my password and verifying my e-mail.

        I don’t “mistrust science,” I was just pointing out that the region had produced one hoax, so obvious they couldn’t get away with it, but it got put into Nat Geo by a respected member of the relevant branch of science. So in this particular area, how strong is peer review going to be if the hoax is really good? What other fossils do you know of from recent history that are celebrated and made the cornerstone of an important theory (or sub-theory?) that come from an anonymous amateur’s work on an unknown site without any documentation? No, I don’t distrust science when it follows proper procedure, but when part of it takes shortcuts and that’s accompanied by spectacular finds that aren’t verified by similar discoveries elsewhere? Well, I can’t help being a bit suspicious, okay?

        As for evolution being the best explanation, current scientific dogma says that everything can and must be explained by natural processes, which doesn’t allow for anything BUT evolution as a viable explanation. However, if you look at the known mechanisms capable of producing biological change and the observed examples of biological change, there’s no indication that such processes could produce all forms of life from single-celled organisms (or powered flight starting with non-flying creatures) over any number of generations or amount of time.

        • Smartass

          Hi David,
          We have to rely on scientists’ vigilance to prevent incorrect conclusions spreading from unverified discoveries. Scientists (I imagine) don’t like wasting their lives following incorrect conclusions, so there is a clear incentive for them to be appropriately suspicious in these cases.
          Regarding your point about the extent of biological changes, I wonder what aspect of evolution could possibly stop small changes adding up to make large ones. Why would there be a limit?


            Who watches the watchmen? Scientists have not lost the human tendency to be more accepting of things that fulfill their wishes than things that would undermine their beliefs. That they even give consideration to such undocumented fossils shows this. Even if the fossils are real, then, how can we trust the statements of belief about all the imagined forms and inter-relationships?

            Small changes can’t add up to all kinds of big ones the way grains of sand make a desert. Big changes can require totally different phenomena as causes. A turtle may climb a hill because it slopes gently, but if you see a row of turtles on fenceposts, you know they didnt climb them.

          • Smartass

            Of course scientists have the fallibility you described: they have the whole list of fallibilities like the rest of us. But science is good at being more reliable as a whole than the individual scientists. It’s demonstrated this well enough over the years.

            It’s reasonable to trust a statement about imagined forms if it follows from an explanation of evidence that becomes widely accepted. That’s always been the way science has discovered things that can’t be perceived by familiar means, from electrons to black holes.

            I asked why small changes couldn’t add up to big ones and you have just repeated the assertion that they can’t, without saying why. The analogy of the turtles on fenceposts sounds like a statement that “irreducible complexity” exists, but this needs justification to be taken seriously. When we look at features that seem at first sight to be irreducibly complex, they turn out not to be, so it seems that nature’s turtles are standing on hills rather than fenceposts.

          • David Bump

            Yes, when the good ol’ scientific method is applied, the influences of human bias and error can be held to a minimum; however, when “science” becomes speculative statements of what might have been, and it has not even been demonstrated that such things can indeed be produced by the mechanisms suggested (or vaguely hinted at), the only real limit is the personal preferences of the majority. “Becoming widely accepted” may amount to nothing more than winning a popularity contest.

            I thought I had explained why small changes can’t add up to big changes. I don’t know about “irreducible complexity,” but I know I haven’t heard of any biological process with the power to increase complexity anything like what is required. The examples I’ve seen of “features that seem at first sight to be irreducibly complex” but supposedly “turn out not to be,” have all required series of significant changes in interactive parts and functioning which haven’t been matched by any observed natural biological variations I am aware of. The slopes of the “hills” seem to be composed of fancy, intelligently imagined stories. I’m sure if there had been such observations, they would be quite well publicized.

          • Smartass

            Hi David, thanks for replying.
            A popularity contest? Fair enough. I’m happy with the idea of taking popularity as a positive indicator when choosing whether to believe a theory.
            I still don’t understand what made you adopt the position that small changes can’t add up to big changes – a position that seems extraordinary to me.
            You seem to assume that, in order for small changes to accumulate into big ones, some of them must individually be large (“increase complexity anything like what is required”). Why would that be? What’s wrong with using lots of individual changes that increase complexity just a tiny bit each time?
            I wouldn’t expect you or anyone else to have seen all the natural biological variations that have ever happened so I don’t know why you think it’s a problem for evolution if it requires variations you haven’t seen.
            Your perception of intelligently imagined storeys in the hills doesn’t sound like a reason to discount evolution since it’s well-known that evolution produces an illusion of design.

    • Jon Tennant

      Hi David, ‘Smartass’ responded with most of what I’d have said about this.

      Yep, Archaeraptor was a chimera and a hoax. That is one among thousands of fossils being unearthed from China atm. These one’s are all real – you can actually check the sediment, and use infra-red techniques to check for validity. The hundreds of scientists and museum curators who seemlessly check these fossils and each others work can’t all be wrong!

      So faith is belief without evidence. Over the last 1-15 years, the evidence coming from China, Mongolia, and other places that track the lineage of maniraptoran theropods through to birds is insurmountable. It changes in nature due to the very nature of science – the things we see, the little details on dinosaurs, can be interpreted differently by different authors. However, these changes are largely only to do with changing small parts of the tree – for instance, in this case shifting the position of Archaeopteryx due to the wealth of new information described by the team. Things which don’t happen as often are the quite large shifts for species like Balaur. Scientists can miss things, we’re not perfect, and that’s why we go over our results, check them, and check again. As the evidence is reviewed by fresh eyes, new conclusions can be drawn based on new evidence – that’s how this stuff works. As Gareth said, this is unlikely to be the final point in this story, but the general trend is still established – who the main players in the story are though, only time will tell!

      • David Bump

        Sorry I’m so late, been busy IRL and had major problems with my password and verifying my e-mail. Likewise, see my reply to “smartass”.

        Yes, Jon, all those people CAN be wrong. For one thing, not that many actually do all that checking on each fossil, and as any good magician or military tactician will tell you, all the assurances you can get are only good enough to cover what happened last time; you never know what someone may have come up with since then. If one region about the size of Michigan keeps turning up just what researchers are hoping to find next, and no scientists are seeing so many of these beauties still in the ground, isn’t that cause to wonder? Suppose instead of dino-birds, it was human remains, but still all these checks of the matrix were coming up valid; would you accept them, or say there MUST be some new way to fake the fossils going on, might the lack of scientific documentation of their removal clinch your belief they should be totally rejected out of hand? But if instead of being “impossible” fossils that would disprove evolution, what if they were merely modern bird forms that would be very hard to explain without a really extreme case of ghost lineage? Maybe some concern? But, when the fossils are merely just exactly what a lot of people want to see… no doubt at all?

        There’s actually some problems with the tracking of maniraptoran theropods through to birds. The main thing is simply that we’ve never been able to demonstrate that the necessary series of transforming and accumulating mutations ever happens. And the fossil record doesn’t actually show a progression from scaly reptilian ancestors to modern birds. A lot depends on rearranging the examples from the series produced by their assigned dates over tens of millions of years, and making assumptions about common ancestry producing features on fossils which don’t actually display them, and other such data-shuffling explanations. What other area of science depends on playing three-card monte with the evidence? I don’t trust the establishment of trends when the trend is highly desired and the watchmen are hoping for pretty much the same thing as the ones they are supposed to be watching, and they validate data that they wouldn’t even consider if it supported something they didn’t want.

        • Jon Tennant

          No probs, hope things IRL are ok! Much of what you state here is speculation and a bit straw man-y, though – sorry.

          There is an excellent documentation now of the incremental changes through the maniraptoran lineage (I can email you the article this is based on if you’d like – you can catch me at @protohedgehog on Twitter). The fact is that the stratigraphical appearances of fossils, both old and new, largely conform to where we expect them to be. This isn’t based on a superficial observation of the fossils, but in depth study. Andrea Cau has analysed many of these fossils first hand over many years, that’s why the analysis consists of such a large matrix. The position of Balaur came as a surprise, and when I rang Gareth he was actually in Romania checking out the fossil again to double check, seeing as the odd placement was so surprising!

          As I mention in the article too, this is hardly the last word. The lineage transformations will become more apparent as new fossils turn up and the tree becomes more dense. This is how science works – the current hypothesis is rigorously based and tested, but subject to change if new evidence should cause it to do so. We can’t see mutations, you’re right (I assume you mean genetic ones), but we do see character state changes (particular aspects of an aspect of morphology) which is how we reconstruct the relationships between species.

          It’s not data shuffling – heuristic and well-understood methods are applied that describe the ways in which character states transform through lineages (cladistics).

          A response to this paper did come out shortly after, simply questioning the provenance of the specimen. So scientists are being vocal about this, and checking and re-checking the work, although the response does little to change the hypotheses outlined in the original article.




            Yes, in addition to not having access to any of the fossils, I haven’t been able to keep up with developments as I would like, so naturally I can only speak in cautious, hypothetical terms, “speculative and straw-many” if you insist. I was just looking at the supplementary info again last nigh, though, and I was surprised to see how far apart some fossils were placed and how some that seem quite different were close together. And there are still late Cretaceous fossils close to the basal nodes while Aurornis, Archie, and other Jurassic and early Cretaceous birds dominate the latter branches, and right at the end there’s a rapid advance (re: flight adaptations) in form but still early Cretaceous, then a jump or rapid series to very modern-like, very late Cretaceous Gansus.

            I’m afraid “heuristic and well-understood methods are applied that describe the ways in which character states transform through lineages (cladistics)” could all-too-easily be a jargon smokescreen for data shuffling.

    • Thomas Diehl

      If there was any difference between a flying feathered dinosaur and a bird, I might consider your point. However, there isn’t. Yes, birds have a reduced tail and lack teeth, but then again so do some dinosaurs. As for Archaeoraptor, that one is interesting: While Archaeoraptor itself was a fake, it was made out of two real fossils, one of which turned out to be Microraptor, an actual flying feathered dinosaur.

      • David Bump

        I never thought the “birds have a reduced tail and lack teeth” argument was all that relevant. If you look at the 40 or so orders (?) of modern birds, they are all quite distinct even from the Cretaceous birds. From what I have read, they also show up fairly early in the Cenozoic, already with their distinctive features. Likewise with bats, btw. So I would also call flying feathered dinosaurs “birds,” but I think they were a distinct form or group of forms. I look at a number of features especially related to flight, consider living flightless birds and the difficulty of developing flight compared to losing it, resist the concept that “ghost lineages” can explain any number of “primitive” forms turning up only after the “advanced” ones, and conclude that the most straightforward interpretation of the fossil record is that flying forms were (generally, more likely) the ancestors of the flightless feathered ones, if indeed all the ones being considered feathered did indeed have some.

  • Chris Brantley

    The dinosaur/bird situation is so fluid, new species of feathered dinosaurs are being found so often, that what we know as the “base avian ancestry” could change next week or even tomorrow. I mean it’s ridiculously premature to form any sort of conclusion until a great many fossils are found. Speculate and theorize all you want but i’d seriously hold off on the conclusions or consider anything “settled”.
    I will say this – Archaeopteryx has fully developed, modern flight feathers. I believe there is an ancestral Great-Grandarchaeopteryx out there with fully developed feathers too. If not, then Archaeopteryx must still lay claim to some sort of oldest bird title because it flew (perhaps in an underpowered way but it flew).

  • Francesco Sinibaldi

    La mémoire


    et comme

    le chant d’une

    pensée, le

    tendre oiseau

    retrouve le

    sourire de





Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar