Return to article page

To print: Select File and then Print from your browser's menu.
This story was printed from FindArticles.com, located at http://www.findarticles.com.

Natural History
Nov, 2000

Tales of a Feathered Tail.(examination of avian evolution from dinosaurs and characteristics of evolutionary progression)

Author/s: Stephen Jay Gould

Soon after birds branched off from dinosaurs and gained the power of flight, some may have branched off again as land-bound runners. Why do some people have a problem with that?

One fine day, or so the legend proclaims, Joseph Stalin received a telegram from his exiled archrival, Leon Trotsky. Overjoyed by the apparent content, Stalin rounded up the citizenry of Moscow for an impromptu rally in Red Square. He then addressed the crowd below: "I have received the following message of contrition from Comrade Trotsky, who has obviously been using his Mexican retreat for beneficial reflection: `Comrade Stalin: You are right! I was wrong! You are the leader of the Russian people!'"

But as waves of involuntary applause rolled through the square, a Jewish tailor in the front row--Trotsky's old school chum from yeshiva days--bravely mounted the platform, tapped Stalin on the shoulder, and took the microphone to address the crowd. "Excuse me, Comrade Stalin," he said. "The words, you got them right; but the meaning, I'm not so sure." Then the tailor read the telegram again, this time with the intended intonation of disgust and the rising inflection of inquiry: `"Comrade Stalin: You are right?? I was wrong?? You are the leader of the Russian people??'"

I have never been able to regard this joke with equanimity, because I can't help wondering what happened to the poor tailor, who probably suffered more than Trotsky, albeit anonymously. But I value the tale as a lesson about the importance of context. We may get every word right but, in a pungent military acronym (the superlative degree of SNAFU, I have been assured by army grammarians), still get the meaning FUBAR (defined by the dictionary, in genteel terms, as "fouled [euphemism] up beyond all recognition"). As I write this penultimate essay of 300 and look back upon a few successes in a sea of persisting ambivalence, I can only conclude that my central subject for "This View of Life"--evolution itself--must surpass all other disciplines in featuring straightforward facts enshrouded in difficult or ambiguous meanings.

The popular understanding of evolution includes at least two false assumptions, so widely shared and so deeply (if unconsciously) embedded in the context of conventional explanations that many plain facts, easily grasped at a superficial level of overt recitation, almost always enter the public discourse of newspapers, films, and magazines in a highly confused form that "science writers" either mistake for the actual opinions of scientists or, more cynically, choose to present as the literary equivalent of "easy listening" for succor in drive-time traffic jams.

In the picture conveyed by these two related fallacies, evolution becomes, first of all, the transformation of one kind of entity into another, body and soul. So fish evolve into amphibians in a "conquest" of the land, and apes leave the safety of trees, eventually to become human by facing the dangers of terra firma with a weapon in a liberated hand and a fresh twinkle of insight behind the eye. In the second component of this transformational view, descendants win victory from the heart of their valor in the face of natural selection--for "later" must mean "better," as the land yields to explorational metaphors of conquest or colonization while the African savannas, for the first time in planetary history, hear sounds of progress in the voice of real language.

But evolution proceeds by the branching of bushes, not by the morphing of one form into another, with the old disappearing into the triumph of the new. Novelties begin as little branches on old trees, not as butterflies of Michael Jordan refashioned from the caterpillar components of Joe Airball. Moreover, most novelties, at least at their origin, grow as tiny twigs of addition to persisting and vigorous bushes, not as higher realizations of ancestors that literally gave their all to a transcendence of their former grubby selves.

Amphibians and all their descendants have done well enough on land, but fins beat feet on the vertebrate bush, where the majority of twigs (species) sprout among fishes. I do not deny the transient success, and interesting novelties, of humans. But Homo sapiens occupies only one twig on a modest primate bush of some 200 species, and even our most distantly related subgroups, in both evolutionary and geographic terms (say, the San of southern Africa and the Sami of northern Finland), show very little genetic divergence, whereas two populations of the same species of chimpanzee, separated by only a few hundred miles of African real estate, have evolved many more genetic differences, one from the other. (This initially surprising fact makes evident sense once we recast our conceptions in properly bushy terms. All living humans descended from common ancestors who lived in Africa less than 200,000 years ago--despite our subsequent spread throughout the world. The two chimpanzee populations may have remained in geographical proximity, but they split from a common ancestor far longer ago, thus providing much more time for the evolution of genetic differences in the separated groups.)

Finally, and at the broadest scale, we will grasp the principle that novelty arises by branching and not by the wholesale transformation of all ancestors into better descendants only when we recognize that bacteria still constitute most of life's tree--including the entire basal trunk that they built by themselves at life's cellular origin--and that all multicellular kingdoms occupy just a few, if admittedly quite healthy, branches at the terminus of a single bough.

During the quarter century that I have been writing these columns, I have continually returned to this theme of mentally liberating bushes versus constraining ladders because I believe that no other misconception so skews public understanding of evolution. I have treated a variety of topics under this rubric: why the air bladders of fishes evolved from lungs and not vice versa, as nearly everyone assumes (including Darwin himself in this case); why the cramming of primates into a halfway corner, and not at a triumphant terminus, of a linear walk through this Museum's hall of fossil mammals makes such revolutionary sense; and why the "out of Africa" theory (on the origin of all modern humans from a recent population of African ancestors) and not the multiregional theory (of our threefold parallel origin from ancestral Homo erectus populations in Europe, Africa, and Asia) represents conventional evolutionary thought based on origin by branching and not the iconoclastic shock featured in most press reports, which have also misconstrued the truly peculiar and theoretically unlikely multiregional theory as transformational orthodoxy.

May I therefore venture one last time into this particular breach (before writing my Agincourt of next month's true finale)? I have desisted in applying this favorite theme to serious public misunderstandings of the apparently accurate claim that birds descended from dinosaurs--probably because I don't like to attack generalities head-on but prefer the path of insinuation by small but fascinating tidbits and also because dinosaurs really are just a tad overexposed and scarcely need more publicity from this student of snails. But my tidbit arrived just in time in the professional literature, thus permitting a swan song about the bushy reform of avian origins at two levels: first the dreaded generality and then the tidbit.

1. The basic relationship of birds and dinosaurs. I don't mean to toss any cold water upon the almost surely valid claim, and one of the most interesting conclusions of late-twentieth-century paleontology, that birds descended from a lineage of small-bodied, bipedal dinosaurs. But the conventional interpretive "take" on this accurately stated fact could benefit from a flat-out dousing, if only because the gain in general understanding of evolution might more than compensate the loss of a charming but truly misleading characterization of a fact.

I should point out, first of all, that the basic claim does not justify the feelings of surprise or weirdness conveyed by most popular accounts. Birds did not evolve from massive sauropods or antediluvian, tanklike ankylosaurs or even from the large tyrannosaurs (which do, in fact, branch fairly close to birds on the dinosaur bush). Rather, birds branched off from a lineage of small, two-legged, meat-eating running dinosaurs--full members of the group by proper criteria of descent (hence the validity of the one-liner that "birds evolved from dinosaurs") but scarcely a version calculated to evoke either the fear or the power associated with our usual icon of dinosaurian immensity.

Moreover, the assertion of this evolutionary linkage during the past twenty years does not mark a stunningly new and utterly surprising discovery but rather reaffirms an old idea that seemed patently obvious to many paleontologists in Darwin's day (notably to T. H. Huxley, who defended the argument in several publications) but then fell from popularity for a good reason based on an honest error.

The detailed anatomical correspondence between Archaeopteryx, the earliest toothed bird of Late Jurassic times, and the small running dinosaurs now known as maniraptors (and including such popularized forms as Deinonychus and Troodon) can hardly fail to suggest a close genealogical relationship. But following Huxley's initial assertion of an evolutionary link, paleontologists reached the false conclusion that all dinosaurs had lost their clavicles, or collarbones--a prominent component of bird skeletons, where the clavicles enlarge and fuse to form the furcula, or wishbone. Since complex anatomical structures, coded by numerous genes working through intricate pathways of development, cannot reevolve after a complete loss, the apparent absence of clavicles in dinosaurs seemed to preclude a directly ancestral status to birds, though most paleontologists continued to assert a relationship of close cousinhood.

The recent discovery of clavicles in several dinosaurs--including the forms closest to birds--immediately reinstated Huxley's old hypothesis of direct evolutionary descent. I don't mean to downplay the significance of a firm resolution for the evolutionary relationship between birds and dinosaurs, but for sheer psychological punch, the revivification of an old and eminently sensible idea cannot match the impact of discovering a truly pristine and almost shockingly unexpected item in nature's factual arsenal.

But I throw my iciest pitcher of water (a device to open eyes wide shut, not an intended punishment) into the face of a foolish claim almost invariably featured--usually as a headline--in any popular article on the origin of birds: "Dinosaurs didn't die out after all; they remain among us still, more numerous than ever, but now twittering in trees rather than eating lawyers off john seats."

This knee-jerk formulation sounds right in the superficial sense that buttresses most misunderstandings in science, for the majority of our errors reflect false conventionalities of hidebound thinking (conceptual locks) rather than failures to find the information (factual lacks) that could resolve an issue in purely observational terms. After all, if birds evolved from dinosaurs (as they did) and if all remaining dinosaurs perished in a mass extinction triggered by an impacting comet or asteroid 65 million years ago--well, then, we must have been wrong about dinosaurian death and incompetence, for our latter-day tyrannosaurs in the trees continually chirp the New Age message of Jurassic Park: life finds a way. (In fact, as I write this paragraph, a mourning dove mocks my mammalian pretensions in a minor key, from a nest beneath my air conditioner. Sic non transit gloria mundi!)

Only our largely unconscious bias for conceiving evolution as a total transformation of one entity into a new and improved model could buttress the common belief that canonical dinosaurs--the really big guys, in all their brontosaurian bulk or tyrannosaurian terror--live on as hawks and hummingbirds. For we do understand that most species of dinosaurs just died, plain and simple, without leaving direct descendants. Under a transformational model, however, any ancestral bird carries the legacy of all dinosaurs at the heart of its courageous persistence, just as the baton in a relay race embodies all the efforts of those who ran before.

However, under a corrected branching model of evolution, birds didn't descend from a mystical totality but only from the particular little bough that generated an actual avian branch. The dinosaurian ancestors of birds lie among the smallest bipedal carnivores (think of little Compsognathus, tragically mistaken for a cute pet in the sequel to Jurassic Park)--creatures that may be "all dinosaur" by genealogy but that do not seem incongruous as progenitors of birds. Ducks as direct descendants of Diplodocus (a sauropod dinosaur of maximal length) would strain my credulity, but ostriches as later offshoots from a dinosaurian line that began with little Oviraptor (a small, lithe carnivore of less than human, and much less than ostrich, height) hardly strains my limited imagination.

As a second clarification offered by the branching model of evolution, we must distinguish similarity of form from continuity in descent: two important concepts of very different meaning and far too frequent confusion. The fact of avian descent from dinosaurs (continuity) does not imply the persistence of dinosaurs (similarity in form and function). Evolution does mean change, after all, and our linguistic conventions honor the results of sufficiently extensive changes with new names. I don't call my dainty poodle a wolf or my car a horse-drawn carriage, despite the undoubted ties of genealogical continuity.

To draw a more complex but precise analogy in evolutionary terms: Mammals evolved from pelycosaurs, the "popular" group of sail-backed reptiles often mistaken for dinosaurs in series of stamps or sets of plastic monsters from the past. But I would never make the mistake of claiming that Dimetrodon (the most familiar and carnivorous of pelycosaurian reptiles) still exists because I am now typing its name, while whales swim in the sea and mice munch in my kitchen. In their descent from pelycosaurs, mammals evolved into such different creatures that the ancestral name, defined for a particular set of anatomical forms and functions, no longer describes the altered descendants. Moreover, and to reemphasize the theme of branching, pelycosaurs included three major subgroups, only two bearing sails on their backs. Mammals probably evolved as a branch of the third, sailless group. So even if we erroneously stated that pelycosaurs still lived because mammals now exist, we could not grant this status to a canonical sail-backed form, any more than we could argue for brontosaurian persistence because birds descended from an entirely unrelated lineage of dinosaurs.

2. A tidbit with feathers. If birds evolved from small running dinosaurs, and if feathers could provide no aerodynamic benefit in an initial state of rudimentary size and limited distribution over the body, then feathers (which, by long-standing professional consensus and clear factual documentation, evolved from reptilian scales) must have performed some other function at their first appearance. A thermodynamic role has long been favored for the first feathers on the small-bodied and highly active ancestors of birds. Therefore, despite some initial skepticism, abetted by a few outlandish and speculative reconstructions in popular films and fiction, the hypothesis of feathered dinosaurs as avian ancestors gained considerable favor. Then, in June 1998, Ji Qiang and three North American and Chinese colleagues reported the discovery of two feathered dinosaurs from Late Jurassic or Early Cretaceous rocks in China ("Two Feathered Dinosaurs from Northeastern China," Nature 393, June 25, 1998).

The subject has since exploded in both discovery and controversy, unfortunately intensified by the reality of potential profits previously beyond the contemplation of impoverished Chinese farmers--a touchy situation compounded by the lethal combination of artfully confected hoaxes and enthusiastically wealthy, but scientifically naive, collectors (see Nature 406, August 31, 2000). At least one fake (the so-called Archaeoraptor) has been exposed, to the embarrassment of National Geographic, and many wonderful specimens languish in the vaults of profiteers.

But standards have begun to coagulate, and at least one genus--Caudipteryx ("feather-tailed," by etymology and actuality)--holds undoubted status as a feathered runner that could not fly. And so at least until the initiating tidbit for this essay appeared in the August 17, 2000, issue of Nature, one running dinosaur with utterly unambiguous feathers on its tail and forearms seemed to stand forth as an ensign of Huxley's intellectual triumph and the branching of birds within the evolutionary tree of ground-dwelling dinosaurs. But the new article makes a strong, if unproven, case for an inverted evolutionary sequence, with Caudipteryx interpreted as a descendant of flying birds, secondarily readapted to a running lifestyle on terra firma, and not as a dinosaur in a lineage of exclusively ground-dwelling forms (T. D. Jones, J. O. Farlow, J. A. Ruben, D. M. Henderson, and W. J. Hillenius, "Cursoriality in Bipedal Archosaurs," Nature 406, August 17, 2000).

The case for secondary loss of flight rests upon a set of anatomical features that Caudipteryx shares with modern ground birds that evolved from flying ancestors--a common trend in several independent lineages, including ostriches, rheas, cassowaries, kiwis, moas, and others. By contrast, lineages of exclusively ground-dwelling forms, including all groups of dinosaurs suggested as potential ancestors of birds, evolved different shapes and proportions for the same features. In particular, as the accompanying illustration shows, ground-running and secondarily flightless birds--in comparison with small dinosaurs of fully terrestrial lineages--tend to have relatively shorter tails, relatively longer legs, and a center of gravity located in a more forward (headward) position. By all three criteria, the skeleton of Caudipteryx falls into the domain of flightless birds rather than the space of cursorial (running) dinosaurs.

Jones and colleagues have presented an interesting hypothesis demanding further testing and consideration, but scarcely (by their own acknowledgment) a firm proof or even a compelling probability. Paleontologists have unearthed only a few specimens of Caudipteryx, none complete. Moreover, we do not know the full potential for ranges of anatomical variation in the two relevant lifestyles. Perhaps Caudipteryx belonged to a fully terrestrial lineage of dinosaurs that developed birdlike proportions for reasons unrelated to any needs or actualities of flight.

I do not raise this issue here to vent any preference (for I remain neutral in a debate well beyond my own expertise). Nor do I regard the status of Caudipteryx as crucial to the largely settled question about the dinosaurian ancestry of birds. If Caudipteryx belongs to a fully terrestrial lineage of dinosaurs, then its feathers provide striking confirmation for the hypothesis, well supported by several other arguments, that this defining feature of birds originated in a running ancestor for reasons unrelated to flight. But if Caudipteryx is a secondarily flightless bird, the general hypothesis of dinosaurian ancestry suffers no blow, though Caudipteryx itself loses its potential role as an avian ancestor (while gaining an equally interesting status as the first known bird to renounce flight).

Rather, I mention this tidbit to close my essay because the large volume of press commentary unleashed by the hypothesis of Jones and his colleagues showed me yet again--this time for the microcosm of Caudipteryx rather than for the macrocosm of avian origins in general--just how strongly our transformational biases and our failure to grasp the reality of evolution as a branching bush distort our interpretations of factual claims easily understood by all. In short, I was astonished to note that virtually all these commentaries reported the claim for secondary loss of flight in Caudipteryx as deeply paradoxical and stunningly surprising (even while they noted the factual arguments supporting the claim with accuracy and understanding).

In utter contrast, the hypothesis of secondary loss of flight in Caudipteryx struck me as interesting and eminently worthy of further consideration but also as entirely plausible. After all, numerous lineages of modern birds have lost their ability to fly and have evolved excellent adaptations for running in a rapid and sustained manner on the ground. If flightlessness has evolved in so many independent lineages of modern birds, why should a similar event surprise us merely because it occurred soon after the origin of birds? (I might even speculate that Cretaceous birds exceeded modern birds in potential for evolutionary loss of flight, for birds in the time of Caudipteryx had only recently evolved as flying forms from running ancestors. Perhaps these early birds still retained enough features of their terrestrial ancestry to facilitate a readaptation to ground life in appropriate ecological circumstances.) Moreover, on the question of timing in our admittedly spotty fossil record, Archaeopteryx (the first known bird) lived in Late Jurassic times, while Caudipteryx probably arose at the beginning of the subsequent Cretaceous period--plenty of time for a flying lineage to redeploy one of its species as a ground-dwelling branch.

After considerable puzzlement, I think that I finally understand the reason for such a stark contrast between my lack of surprise and the sense of deep paradox conveyed by most press reports. As an implication of my view (expressing a professional consensus) that evolutionary novelty arises by a process of branching, the discovery of an earlier "first time" for a common and repeated event--loss of flight and secondary adaptation to effective ground running--surely attracts interest as a lovely nugget of discovery but scarcely evokes any theoretical surprise.

But in the usual public misconception of evolution as a story of wholesale transformation into something better, such an early "falling off" from "the program" seems almost perverse. After all, birds had just taken to the air a few tens of millions of years (at most) before the appearance of Caudipteryx. Why would a lineage fall out of step so early in the game? Sure, once the program rolls to a full and triumphant completion, then evolution might permit an ostrich or two to slip off the main line and pursue its own bohemian path in a now strange but once ancestral land. But such events surely cannot occur in the vigorous youth of a lineage that has just snatched winged victory from the jaws of terrestrial dinosaurian death.

Perhaps I have treated a garden-variety error with unfair disdain in the sarcasm of the last paragraph. But the fallacy behind this common feeling of surprise, evoked by the eminently plausible hypothesis of Caudipteryx as a flightless bird, originates in a pervasive bias that renders much of the fascination of evolution inaccessible to millions of genuinely interested students and lovers of science.

The vigorous branching of life's tree, and not the accumulating valor of mythical marches to progress, lies behind the persistence and expansion of organic diversity in our tough and constantly stressful world. And if we do not grasp the fundamental nature of branching as the key to life's passage across the geological stage, we will never understand evolution aright. Tennyson caught the essence of life's challenge when he personified nature's unforgiving geological ways, as expressed in the fossil record of extinction:

   From scarped cliff and quarried stone

   She cries, "A thousand types are gone:

   I care for nothing, all shall go."

Yes, all shall eventually go, but some shall branch, thus permitting life to persist. To cite a sardonic song of self-mockery in leftist circles: "Trotsky got the ice pick ... and so say all of us." And I do shudder even to contemplate the fate of the poor tailor. But life feints, dodges, and branches--and therefore, above all, hangs on in beauty and fascination. Psalm 1 invokes the right picture for a different purpose: "And he shall be like a tree planted by the rivers of water ... its leaf also shall not wither; and whatsoever he doeth shall prosper." And Darwin employed the same image, both as metaphor and as literal topology this time, in the final words of the focal chapter in the Origin of Species--chapter 4, entitled "Natural Selection," with its closing literary flourish on extinction and branching as the motors of evolution's tree and life's glory:

   As buds give rise by growth to fresh buds and these, if vigorous, branch
   out and overtop on all sides many a feebler branch, so by generation I
   believe it has been with the great Tree of Lift, which fills with its dead
   and broken branches the crust of the earth, and covers the surface with its
   ever branching and beautiful ramifications.

Stephen Jay Gould teaches biology, geology, and the history of science at Harvard University. He is also Frederick P. Rose Honorary Curator in Invertebrates at the American Museum of Natural History.

COPYRIGHT 2000 American Museum of Natural History
in association with The Gale Group and LookSmart. COPYRIGHT 2000 Gale Group