The discovery of Archaeopteryx 150 years ago provided elegant (and well-timed) support for Darwin’s theory of evolution, laid out in detail only two years earlier in his 1859 publication of On the Origin of Species . The exquisitely preserved plumage of Archaeopteryx, combined with its toothy beak, long bony tail and fingered hands marked the animal as a clear example of an evolutionary intermediate between reptiles and birds . This was especially vindicating in Darwin’s day because the fossil record then was still relatively devoid of specimens predicted by evolution to bridge the morphological gap between major groups (e.g. birds) and their alleged ancestors (e.g. reptiles). In fact, for 150 years, Archaeopteryx has been considered by many people to be the first bird. What a title to be expected to uphold! Well, today, the discovery of the small feathered dinosaur (Xiaotingia) is causing a great flap in the media because a reanalysis of the relationships among dinosaurs and birds prompted by this discovery suggests that Archaeopteryx is not a bird after all, but rather a small feathered dinosaur . But should feathers fly over Archaeopteryx‘s hop to a nearby branch of the tree of life?
First, a bit of background and applause for the very unfortunately overshadowed discovery that sparked the collective media gasp. Xiaotingia zhengi was a small crow-sized dinosaur with long, graceful arms that were equipped with both feathers and clawed fingers . The fossil’s marvelous state of preservation is characteristic of many Chinese specimens that are helping paleontologists trace out the early evolution of birds. Xiaotingia strongly resembled its (currently) nearest known relative, Anchiornis huxleyi, also from late Jurassic China. Both species are closely related to Archaeopteryx lithographica. The news that is making waves comes from the redrawn family tree of Archaeopteryx among birds and dinosaurs once the relateness of Xiaotingia is also considered. Under these conditions, Archaeopteryx no longer perches on the branch in common with the lineage of modern birds but rather on the branch of the deinonychosaurs, which include dinosaurs such as Velociraptor. It appears that Archaeopteryx was not the ‘first bird’, an unknown animal to which other feathered fiends (e.g. Epidexipteryx, Jeholornis or Sapeornis) now appear closer related. But the popularity of this news suggests that the fall of Archaeopteryx from its perch is a lot more important than it really is.
Why am I raising a stink about the outcry over Archaeopteryx‘s fall from grace? As a microbiologist, I’ve seen this kind of rearrangement of classification happen time and again with the bacteria that I study. I merely take it with a shrug, or occasionally a sigh when it causes me more work to account for it in the discussion section of the scientific paper describing a species. But Archaeopteryx commands a kind of public awe that the microbial subjects of my research could only envy, had they the means. In particular, Archaeopteryx holds a near-iconic position as an example of an evolutionary intermediate, the ‘first bird’, not yet even free of its reptilian bonds. Should any discovery threaten to shoo the first bird from its perch, proponents of creationism are anticipated to rush in headlong to claim that evolutionary biologists had misinterpreted the fossil, that it was not an intermediate between animal ‘kinds’ at all. True, this discovery changes things a bit, but not in the way that creationists would have us think, and certainly not in a way that threatens to topple the very foundations of evolution. Allow me to explain.
Creationists unreasonably demand evolutionary biologists and paleontologists to supply evidence for evolution in the form of “transitional forms”. These organisms are supposed to provide a bridge between living animal “kinds” (as defined by the Bible, which, by the way, does not elaborate on the precise meaning of the word). Their assertion is that god created different “kinds” of animals (we might assume that birds and reptiles belong to different “kinds”), and that one “kind” does not evolve into another. But it is flawed reasoning to require that we find an animal that is half way between one major taxonomic (i.e. related by classification) group (e.g. birds, class Aves, or slightly more inclusively, Avialae), and another (e.g. reptiles, class Reptilia).
It’s useful to think of today’s well-separated classes of vertebrates like the islands of a volcanic archipelago: they arose from one source of upwelling magma, but as it rose, the source branched, forming several seamounts; the ocean now separating them hides the connectivity that they possessed in the deep sea, in deep time before they recently reached the surface. Similarly, it’s unreasonable to expect to find something half way between major animal groups after they have changed so much from their common ancestor. Extinction has hidden a lot of detail of the path of evolution leading to modern groups. Rather, we should expect to find fossils of ancestors of both lineages that increasingly resemble each other the further back in time that we look, but not necessarily resembling either major living group. We are really searching for a common ancestor, not a mythical crocoduck.
This brings up another problem. Given how difficult it is for a dead thing to become fossilized (a lot of conditions need to be met to preserve its remains for millions of years), we cannot expect to find every link in the chain of descendents between a modern species and a distant ancestor. The best we can hope for is a string of highly disjointed dots. So how do we really know whether we have ever recovered a true common ancestor of two groups at all? We don’t. In fact, it is astronomically more likely that what any presumed evolutionary sequence of fossils represents is not a direct line of ancestry or a geneology, but rather a series of species each of which has branched off and evolved independently a short distance from the line of ancestry that we seek. Finding these fossils is analogous to walking through an autumn forest and picking up leaves that fell from tips of tree branches rather than from the main trunk (let’s pretend that trees can bear leaves directly on their trunks and on their branches, for the purpose of argument). Trees have a lot more branches than main stems or trunks, and therefore it is muchmore likely that if we pick up a leaf at random, it will have come from a side branch somewhere. We will probably never hold the fossil of the real ‘first bird’.
But cry no tears, for this doesn’t matter. We can still reasonably reconstruct an evolutionary sequence, and guess very well at the appearance of a hypothetical common ancestor, using only the remains of species that have evolved a small amount since branching from the common ancestor or the line of ancestry. All we need to do is look at the characteristics that they share (those characteristics that have not evolved significantly since the branch point). The last common ancestor between two species is likely to have possessed those characteristics that they still share. For example, we can hypothesize that the common ancestor of birds and closely related dinosaurs was also feathered, because there exist fossils of both early birds and undisputed dinosaurs that possess feathers. Moreover, if we have a whole series of fossils that may roughly delineate an evolutionary sequence (subject to the constraints of interpretation that I have just highlighted), then we can measure a whole bunch of their characteristics (wing length, beak length, presence or absence of claws on their hands, etc., etc.) and arrange them according to a hierarchy of most to least shared characteristics. This is the purvue of taxonomy and cladistics, the science of determining the relatedness of organisms on the basis of their physical characteristics. Arranging the fossils according to the hierarchy of their shared characteristics allows taxonomists to generate a ‘family tree’ of sorts (which they call a cladogram), outlining the order in which the species are likely to have descended along one or more lineages. Even though Archaeopteryx is not likely to be the common ancestor between birds and related dinosaurs, we are lucky to have uncovered it, for it nevertheless lies very near the common ancestor. And it is the proximity of Archaeopteryx to the hypothetical ‘first bird’ that has caused its classification to change.
But the effective change in the classification of Archaeopteryx as a result of the discovery of Xiaotingia is not large anyway. The nearer in time that we approach the evolutionary ancestor of birds and reptiles, the less meaning there is in the organizational category of “class” (using the taxonomic definition of the word, of which birds and reptiles each constitute a unit). If a wall is painted red on one end and gradually fades to blue on the other, then it is more difficult to categorize a section near the purple center as more red or more blue. The distinction loses practical meaning. Archaeopteryx is so near the common ancestor of birds and related reptiles that it takes only relatively small changes in the proportions or positions of its bones to toggle the animal between the bird and reptile classes. In fact, it is more useful and precise to compare Archaeopteryx not to members of Reptilia in general but rather to the specific reptile group to which it was most closely related, the small theropod dinosaurs known as the Deinonychosauria (known to most by the popularized term ‘raptors’).
Furthermore, it is a phenomenon very familiar to taxonomists (scientists who classify and name life forms) that the inclusion of additional species into a pool of species that are to be arranged into a hierarchy of relatedness (e.g. Kingdom > Phylum > Class > Order > Family > Genus > Species) can change the topology (the arrangement of connections) of the resulting cladogram (‘family tree’). This phenomenon is an unavoidable result of the statistical techniques that are used to calculate the most parsimonious (reasonable or simple) manner of branching of the tree of relatedness. It’s especially likely to happen when a new species is added that has features in common with two different groups. This new species exerts a kind of ‘mathematical gravity’ on other species that originally fell near the borderline between major groups. Such a fate has befallen poor Archaeopteryx. The inclusion of Xiaotingia into the mathematical analysis has plucked Archaeopteryx from Avialae and dragged it the very short distance into the neighboring Deinonychosauria on the basis of their shared characteristics. That’s all.
Archaeopteryx is no less a bird morphologically than it was before. The analysis that dethroned it was not a description of previously hidden anatomical features of the animal (e.g. milk ducts). As Witmer (2011) noted, Archaeopteryx still has the birdlike features of feathers, furcula (wishbone), three-fingered hands, long arms and retroverted pubis (backward-pointed hip bone) . And, via the same type of cladistic analysis that removed Archaeopteryx from the bird class, future discoveries may bring Archaeopteryx back into the flock of Avialae once again.
So this is certainly not the end of the story. The point is that the inclusion of Archaeopteryx in either birds or reptiles is not as significant a point as it first seems because birds and the most closely related reptiles were a lot more similar in the days of Archaeopteryx than they are today. Furthermore, demonstrating the evolution of birds from dinosaurs, as we have seen, does not require possession of the actual ‘first bird’, whatever that may be.
1. Darwin, C. (1859) On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life (1st ed.) London: John Murray.
2. Ostrom, J. H. (1976) Archaeopteryx and the origin of birds. Biological Journal of the Linnean Society 8:91–182.
3. Xu, X., You H., Du K., and Han F. (2011) An Archaeopteryx-like theropod from China and the origin of Avialae. Nature 475:465–470.
4. Witmer, L. M. (2011) An icon knocked from its perch. Nature 475:458–459.