When Did Dinosaurs Learn to Fly?

Birds are dinosaurs. That’s a fact underscored by dozens upon dozens of discoveries in the last 30 years. Free of the historic blinders that cast dinosaurs as monstrous reptiles, we’re now gaining an ever-greater appreciation for how bird-like Tyrannosaurus and its famous relations really were.

In fact, many traits we think as unique to birds evolved hundreds of millions of years ago. Reproduction by laying shelled eggs goes back to some of the first vertebrates to carve out a living on land, around 315 million years ago. Fluffy body coverings might go back to the earliest dinosaurs. And air sacs that radiate out from the respiratory system into bone go back to the last common ancestor of dinosaurs like Apatosaurus and Allosaurus, at least.

But what about flight? More than anything else, the ability to take to the air seems to distinguish birds from most of the extinct dinosaurs, and this is where the picture starts to get a little fuzzy. For symbolic reasons, at least, avian flight marks the arrival of something new and different on the evolutionary scene, and paleontologists have spent over a century trying to tease out the transition.

The latest entry into the field was just published by Yale University researchers Teresa Feo, Daniel Field, and Richard Prum. They focused on one particular part of dinosaurian anatomy – asymmetrical feathers.

The presence of asymmetrical wing feathers – with a short leading edge and longer trailing edge, such as the primaries on the wing – has often been taken as a rough proxy for some kind of flying behavior in extinct, feathered creatures. That’s because this shape helps create lift. As Feo and colleagues point out, though, associating a general shape with the ability to fly is too coarse an interpretation. Many flightless birds have asymmetrical feathers that they inherited from their flying ancestors, including streamlined penguins that flap through the water. In order to tease out the clues of how these feathers contributed to flight, researchers have to comb over the plumage in closer detail.

Feo, Field, and Prum looked to geometry to see how the feathers of non-avian dinosaurs like Microraptor compared to those of early birds, such as Archaeopteryx, and their living relatives. Specifically, the researchers zeroed-in on the angles and lengths of asymmetrical feather barbs – the shafts that run perpendicular to the central rib that the rest of the feather branches out from. Given that the length and the angle of the feather barbs alter flight ability, Feo and coauthors could come up with a better idea of how skilled extinct dinosaurs would have been in the air.

The evolution of primary feather geometry. From Feo et al., 2015.
The evolution of primary feather geometry. From Feo et al., 2015.

In some ways, the primaries of Archaeopteryx and Microraptor were like those of other birds, including living species. The barbs on the “cutting edge” of their feathers were held at small angles relative to the shaft they branched from. This kept the leading edge of the feather relatively rigid and better for pitch control. But, on the trailing edges of their feathers, Archaeopteryx and Microraptor were different from flying birds.

Along the trailing edge of the primaries, Feo and coauthors point out, the barbs of flying birds are positioned at relatively large angles. This helps gives the feathers flexibility and maintain a stable airfoil. But in Archaeopteryx and Microraptor, the trailing-edge barbs were held at small angles. This kept their primaries stiff and less responsive, limiting their degree of flight control.

From their fossil sample, Feo and colleagues hypothesize that “modern” asymmetrical feathers with small leading and trailing barb angles first evolved in early, toothed birds like Confuciusornis and Eopengornis, around 125 million years ago. Along with other traits that evolved around the same time – such as a “winglet” called the alula and expanded bony keel – the barb angles hint that these birds really were flying.

But what about Archaeopteryx and Microraptor? Paleontologists have gone back and forth over whether or not these dinosaurs could fly for years. The emerging consensus is that they were able to move through the air somehow, but perhaps not in a way that would be familiar to us. While they weren’t capable of the “modern avian flight stroke” – the crux of these investigations – Archaeopteryx and Microraptor may have used some combination of gliding and flapping. Watching an airborne Archaeopteryx must have been quite a sight, and, from feather and bone, that is exactly what many paleontologists are trying to envision.


Feo, T., Field, D., Prum. R. 2015. Barb geometry of asymmetrical feathers reveals a transitional morphology in the evolution of avian flight. Proceedings of the Royal Society B. doi: 10.1098.rspb.2014.2864

56 thoughts on “When Did Dinosaurs Learn to Fly?

  1. Dear Brian,

    I must disagree with you on two points. First of all, I do not think that we can classify the primitive Archaeopteryx as a genuine bird. The true Aves, I think, would have appeared later, with Confuciusornis and Iberomesornis. Secondly, birds are not and have never been dinosaurs, any more than mammals are cynodonts or early reptiles amphibians. Birds EVOLVED from theropod dinosaurs, yes; but when the very first bird appeared on this Earth it already belonged to the new Class Aves, it had mutated into a novel, diverse type of creature, and thus could not claim to be a member of the ancestral Superorder Dinosauria.

  2. Birds have feathers of many shapes, from tiny, pan-shaped feathers around their beaks and eyes to symmetrical body/contour feathers, to the symmetrical tailfeathers, etc., etc. What was behind the evolution of these different types of feathers, and how is it that some of them on the “hand” evolved into longer, asymmetrical feathers? What was the original purpose of those longer, asymmetrical “hand” feathers that also served the purpose of some kind of advantageous lift/speed?

  3. Mary,

    In following the old Linnean system of classification, you’re right–birds are no longer dinosaurs in the sense that they are a separate class. However, you must realize how strange that is: how can two orders of equal weight (Reptilia and Aves) exist when one clearly evolved from the other? Paleontologists abandoned Linnean classification as cladistics and phylogenetics arose during the 80’s and was embraced in the 90’s.

    Everything is a subgroup of something else. Birds are a subgroup of maniraptors, which are a subgroup of coelurosaurs, which are a subgroup of tetanurans, which are a subgroup of neotheropods…going back forever. The only Linnean schemes that are still followed today are Family, Genus, and Species for convenience.

    Thus, the placement of Archaeopteryx within Aves is purely subjective, as it currently is in paleontology. You can move the term “Aves” around based on what characteristics you choose to associate with it. Indeed, many workers restrict the term to the crown–living birds and their immediate ancestors. Other prefer to use the term for Archaeopteryx plus all other flighted paravians. The point is, there’s not a definite line in the sand for where “Aves” begins. It’s all subjective, so we’ve stopped using the old system.

  4. Following on from what Zach said, wherever you define the box “Bird” (or “Aves”) to commence, it follows that whichever individual was the first bird, it’s parents were, by your own reasoning, not birds; ie dinosaurs.

    I would wager a large sum of money that you could not discern a meaningful taxonomic difference between this bird and its dinosaur parents, which then begs the question, why is this creature not also a dinosaur?

    The only way that it could not be a dinosaur too, is if the category “bird” had as part of its definition “not a dinosaur”. That artificial construct was overturned more than a quarter of a century ago by the scientists who work in this area.

    1. Dear Zach and Mark,

      I’d like to make two comments concerning what you’ve said. As for the Linnean classification, in my opinion it is a rational, solid manner of determining the placement of living organisms into groups. I do not say that cladistics and phylogenetics are devoid of validity; but since their adoption they have provoked considerable confusion and have hazed the distinctions between life forms which, I believe, for the sake of scientific precision should remain clear.
      According to your own argument, fish, amphibia, reptilia, mammalia should not be able to exist together because of the clear, direct derivation of one order from the other. And yet their separate existence is a reality: I do not see orders as sub-groups which still belong to the ancestral trunk, but rather as separate, new types, possessing particular characteristics which amply differentiate them from their immediate as well as from their more remote forebears.
      As for subjectivity: science, more than all other disciplines, MUST remain objective: otherwise we could assign certain arbitrary characteristics to pteranodons, bats, feathered T-Rexes, who knows, even model airplanes with feathers glued on to them, calling these “birds” or “paravians” according to our personal whim.
      The mutant (i.e. transformed) progeny of an animal no longer belongs to the species, or even order, of its parents if that mutation/evolutionary process is drastic. A cat born with two tails is still a cat; a bird hatched from an egg laid by a paravian or proavian dinosaur is always and forever a bird, BECAUSE ITS GENETIC AND ANATOMICAL STRUCTURE IS SUFFICIENTLY DIFFERENT FROM THAT OF ITS PARENTS TO WARRANT THIS NEW CLASSIFICATION. Certainly there is a meaningful taxonomic difference between this bird and its dinosaur parents, although at first glance this difference might seem very subtle. Palaeontology, with its new sophistication, surely must be able to identify this distinction.

  5. Dear Mary (tempted to write: Dude, Mary)

    You miss the point. Ignore Linnaean classification. Birds are dinosours because they evolved from ancestors which *were* dinosours. Similarly, we (humans) are fish because we evolved from ancestors which *were* fish. Just because we change our “shell” doesn’t mean we change membership.

    More eruditely, you are arguing for “evolutionary classification”, or the mistaken notion that the “gap” or “difference” between organisms merits classification at a different level of the Linnaean hierarchy. Objectively, this is B.S. There is no sound way to quantify the evolutionary gap between lineages; therefore evolutionary classification is in error. Please open your mind. Nature will be way more fascinating once you realize how things are related.


  6. P.S., I can’t help myself, but I will put this out there: Learn tree-thinking. If you don’t know what I mean, ask me about it. Yell if you must, but ask. Again, cheers.

  7. Mary,

    Under your logic, Diplodocus longus, Triceratops horridus, Iguanodon bernissartensis and Stegosaurus stenops are not dinosaurs, since the first dinosaur described (Megalosaurus bucklandii) are more closely related to aves and their ancestors than it is to sauropodmorphs and ornithischians.

    Essentially, as maniraptoran coelurosaurs (and theropods) Deinonychus antirrhopus and Microraptor zhaoianus are much more closely related (in evolutionary terms) to (and share more common features with) Iberomesornis romerali than they are to Ankylosaurus magniventris, so any group that includes maniraptors and ankylosaurs must include birds as well; yet you’d put Deinonychus and Microraptor in the same “group” with Ankylosaurus while leaving Iberomesornis out. To be blunt, your logic makes zero sense whatsoever.

  8. I think it’s a bit silly to say that birds aren’t dinosaurs. Just check out this XKCD comic, http://xkcd.com/1211/. I don’t see how you could make a group that lumped together deinonychus and stegosaurus, but excluded an eagle from that group, when the deinonychus is so much more similar to the eagle. It would be like saying bats or cetaceans were no longer mammals because of how modified they are for their lifestyles.

  9. “…a bird hatched from an egg laid by a paravian or proavian dinosaur is always and forever a bird, BECAUSE ITS GENETIC AND ANATOMICAL STRUCTURE IS SUFFICIENTLY DIFFERENT FROM THAT OF ITS PARENTS TO WARRANT THIS NEW CLASSIFICATION.”

    That’s not how evolution works. Animals always give birth to or lay eggs of animals in their same species. There will almost certainly be a few genetic mutations in the offspring (just like in your own DNA or your children’s DNA), but the changes won’t be enough to keep the new offspring from breeding with other members of their species. I’ve actually written slightly more in depth about this on my own site in an article on which came first, the chicken or the egg – http://www.jefflewis.net/blog/2013/07/which_came_first_the_chicken_o.html

  10. Here’s another way to look at it:

    Separating “birds” from their immediate non-avian ancestors is as arbitrary as separating Pentaceratops from Utahceratops. The differences are so minor that you could easily say that both animals are, practically speaking, Pentaceratops. But what we don’t do is put Utahceratops in a different “suborder” from Pentaceratops.

    There’s nothing inherently special about Mesozoic birds that separates them from non-avian dinosaurs. As I said before, it’s an arbitrary line in the sand, made purely for convenience.

  11. I can just see those scientists in ther white lab coats combing through the dinosaur feathers, while the dinosaurs leaf through thier copy of Dinosaur magazine as they relax in the salon chair.


    It may be true, but you lack these “parent-descendant” pairs, and just have to discuss something scientific based on the actual fossils. And based on the fossils we have, there is NO scientific reason to consider Archaeopteryx something “much different ” from, for example, a Jinfengopteryx to warrant a new classification. Their morphological differences are very few. So, even Jinfengopteryx should be “a bird” as Archaeopteryx. But a Jinfengopteryx is not enough different from a Sinovenator to be classified in a “new class” relative to Sinovenator. And Sinovenator is not much different from a Caudipteryx for the same purpose, and a Caudipteryx is not much different from a Falcarius, and a Falcarius is not much different from an Ornitholestes for the same purpose, and an Ornitholestes is not that much different from a Tanycolagreus, and a Tanycolagreus is not that much different from a Guanlong, and a Guanlong is not that much different from a Dilong, and a Dilong is not that much different from a Yutyrannus, and a Yutyrannus is not that much different from a Tyrannosaurus. So, if you place Archaeopteryx in a new “class” distinct from reptiles, the same must be done for Tyrannosaurus. Otherwise you are contradictiong you claim. The day you’ll find a testable difference between any of the couples of fossils I’ve listed, a difference so large to warrant a “taxonomic distinction” with the magnitude you claim, I will trust your words.

  13. Hello Jeff, Zach, Andrea,

    Sorry to say this, but I think that you’re missing the point. As for plants and animals reproducing their own kind, of course they do, Jeff, UNTIL nature decides that it’s time for a different category of life to appear. If this had never happened, then there would never have been any evolutionary process at all, just one type of creature making perfect carbon copies of itself ad infinitum.

    Zach, Andrea, I have never advocated separating similar animals that obviously belong in one and the same order. Reptiles remain with reptiles and birds of a feather should flock together. I do not believe that science can be arbitrary, failing to recognize that at one particular moment in time the first true mammal, or bird, or any other founder of a new order necessarily appeared: not a borderline case, a transitional example (these are part of a gradual evolutionary process), but a full-fledged member of an order which, let’s admit this, is different in significant aspects from that of its parents. And I am not alone in claiming this. Numerous palaeontologists claim to have pinpointed the “taxonomic distinctions”.

    The first true mammals had already differentiated sufficiently from their reptilian ancestors to be identified as such. The first true birds (and I have been doing years of research to identify them) have a morphology which can no longer be equated to that of dinosaurs. Surely modern palaeontology possesses criteria sufficient to differentiate, classify, determine what a new order really consists of? Archaeopteryx, Aurornis and company may resemble true Aves on the outside, but their anatomical structure will belie this. THEY are still dinosaurs.

  14. Mary, I’m afraid you’re the one missing the point. You’re still approaching evolution from a typological viewpoint, and as a result you’re stuck in a saltational model that simply doesn’t fit reality. A population is composed of organisms that vary (I’m sure you already know this, but it doesn’t seem as if you’ve worked through the implications). The evolution of a population is mostly about shifting the distribution of traits within that population — if it helps, think about a bell-curve that slowly shifts it’s mean from one point to another. Eventually, you shift it far enough that we’re willing to call it something different, but at any point in the process every individual still falls within that curve. Thus, there is no single, definable point at which you can say something new has just showed up.

  15. Mary, I don’t know where you got your ideas about evolution from but I’m afraid it does not work how you seem to think it does. No animals have ever given birth to a completely different animal. the process is considerably more gradual than that.

    From what you wrote, you also seem to misunderstand how sexual reproduction works. Even if evolution stopped occurring, animals would not give birth to “perfect carbon copies” of themselves. That would be parthenogenesis (and even then, gene transfer can occur via a number of different mechanisms such that the progeny will not be exactly the same as the parent).

    You say that you do not believe that science can be arbitrary but, unfortunately, when it comes down to humans putting labels on categories of things in the natural world, that is exactly what it is. Even in physics. As an example, if you look at the spectrum of light, at what frequency does the colour stop being red and become orange?

    It’s the same with animals. If you could be present during the multi-generational process of non-avian dinosaurs evolving into birds, you absolutely would not be able to pick the exact point at which this transition occurred. To that extent, it is arbitrary. That doesn’t mean that it’s a free-for-all and each person gets to decide that point for themselves. Consensus amongst palaeontologists is achieved over many years of study, and is refined as more information becomes available.

    It’s unclear what you mean when you say that [N]umerous palaeontologists claim to have pinpointed the “taxonomic distinctions” but I am not aware of any palaeontologists that would support your views on this.

    Lastly, you say that you have been doing years of research on avian morphology, to identify which creatures are birds. If your research had included the remains of dinosaurs, it would not have escaped your notice that dinosaurs such as Velociraptor and Tyrannosaurus are morphologically more similar to a chicken than they are to either Stegosaurus or Triceratops. Ergo, if we categorise both Triceratops and Tyrannosaurus as “dinosaurs”, we must also include chickens (and, indeed, all birds).

  16. mary, what is your definition of “sufficiently different” and describe how it is not a wholly arbitrary demarcation?

    The anatomical differences between a dolphin and a human are vastly greater than the differences between an eagle and a velociraptor. So what makes the eagle warrant a new class, and not the dolphin?

  17. Mary, as a taxonomist I both sympathize with your position about classification and know the cladists are winning, at least for now.

    When one limb on the phylogenetic tree goes “sideways” to new, significantly adaptations (e.g. flying birds from non-flying dinosaur ancestors), I’m happy classifying them as different, even at the same level of their ancestral group. After all, classification is a human, arbitrary activity, not a true phylogenetic tree. If we don’t do this, we have to treat all mammals, dinosaurs, and birds as particularly odd fish — which is true and yet not a useful classification system.

    Right now, the emphasis is going toward phylogeny-only systems (though the Linnean classification system has not disappeared). In this sense, birds are dinosaurs, dinosaurs are fish, and we must all be classified with the first eukaryotic protistans.

    Despite the fuzziness of where birds start, there is a sense in which birds are different — we can call them birds rather than dinosaurs. And there’s a sense in which they are just really odd dinosaurs. I think both sides should accept the arbitrariness of classification and move on to discover and celebrate the relationships and biology of these organisms.

    1. Hello Barbara,

      Thanks for your comments, which indicate that you are open-minded and fair. Classification is, indeed, a human, arbitrary activity. Not even Linnean classification is immune from errors. I have been studying prehistoric fauna and flora for many years now. I, too, have examined all aspects of palaeontology, both traditional and innovative. It is only after long reflection and consideration that I have opted for the former.

      Palaeontologically speaking, I am essentially and obviously a Linnean traditionalist. I would not continue to be so unless I were convinced of the validity of my own and Linnaeus’ premises. As I AM completely convinced of this, it is perhaps useless to continue my debate with certain other users of this forum who are firmly convinced of the opposite. Your own comments, however, Barbara, are fair and rational; you are willing to consider both sides of the coin, and I appreciate this because you have done so in the impartial spirit of scientific research. You have also been polite to me, where several others have not. Thank you for this.

      You spoke of fuzziness; and I think that this is the essential problem of the new cladistics. Blurred distinctions between life forms, shortage of clear-cut definitions. I do not perceive evolution as long dragged-out continuity, with minor almost imperceptible changes over the generations, but rather as SIGNIFICANT genetic and structural modification with each progressive mutation, no matter how subtle that modification may seem at first examination. I’m not claiming that brand new species arise brusquely from fundamentally different parents. Birds did not appear all at once, magically, from non-avian dinosaurs or even from early, non-derived “paravian” theropods. The development was gradual; but it did result, at least in my humble Linnean opinion, in the advent of a brave new ORDER of animals.

      If I were to look at a pet parakeet and tell him, “You’re a fine little dinosaur!”, he would probably cock his saucy head and stare at me sideways, saying: “A dinosaur? ME?”

  18. The question is not so much whether Aves is considered to be a “Class” but whether birds are dinosaurs. Mary stated categorically that “birds are not and have never been dinosaurs”. This is simply not true.

    Anyone happy to classify birds as being completely distinct from dinosaurs must also do the same for bats and whales and not consider them to be mammals.

    Cladistics is a very useful system and the phylogenetic trees produced depict the relationship of living things in a clear and unambiguous manner. Any abitrariousness over whether to call a particular group a Magnorder or Parvclass is removed. It is these arbitrary labels which are not especially useful.

  19. Mary, gotta love how you’re forced to ignore the latest addressals to your comments. Like the name sake, you ar enothing but a sheep indeed.

  20. Mary, Linnean taxonomy was developed before both fossils and evolutions were known by scientists. It’s a pre-darwinian and strictly-neontological system of classification. This is why it is a too rigid system to be useful in palaeontological evolutionism that is the science studying Archaeopteryx, dinosaurs and bird origin in the fossil record. This is why it may eventually be useful when talking about living beings but produces such absurdities like “birds and reptiles are distinct categories” when is forced to be valid also among the fossil lineages along the geological time.
    You did not understand my comment. My comment said that there is not any large morphological boundary between birds and dinosaurs that you could use to separate the two groups in a Linnean sense. I’ve been studying the bird evolution from other theropods for 15 years, and what I’ve learned is that the more we know of that phenomenon the more we conclude that the boundary between “birds” and “reptiles” is completely arbitrary, non-objective and merely nominalistic, as it is unable to correctly describe the complex branched evolutionary continuum among birds and their ancestors. It is only in the living forms (due to several extinctions that eliminated several morphotypes among intermediate forms leading to living species) that the huge gaps among living forms produces those traditional “bird” and “reptile” distinctions. Also, “reptile” is quite difficult to be defined, as crocodiles are much similar to birds in many relevant features than to lizards…

  21. Geesh, Brian. Just because there is evidence of some feathered dinosaurs, it doesn’t change the fact that there were plenty of truly monstrous, scale covered, and VERY reptilian looking dinosaurs. Crocodilians are archosaurs too, aand considering the fact there are scaly-skin impressions of Albertasaurus, there’s a good chance that its close relative, T-Rex wasn’t as bird-like as you would like to believe. The bottom line is that as surely as there were some dinosaurs that had bird-like characteristics , there were many, many more that had far more reptilian characteristics. Crocodilians and Birds are both archosaurs, and the preponderance of evidence suggest most dinosaurs looked more like crocs than birds, in terms of body coverings. Even among other theropods, like Carnotaurus, we see bony scutes and nodules very much like those of crocodiles. I think most people would say that a scale studded Carnotaurus looks both monstrous and reptilian, and it is highly unlikely it could ever be mistaken for a bird.

  22. Carlos, I’m not trying to ignore anything or to avoid the issue. I’ve simply had my say and will stick to my opinions, no matter what criticism others may choose to express. And to be perfectly honest, it’s offensive comments like yours which make me want to stay away from this forum in future.

  23. Mary, Carlos’ comment aside, I really would like to know how you can justify (even on a personal level) Confuciusornis being in a separate suborder from Archaeopteryx, but Triceratops IS in the same suborder as Archaeopteryx. There are far more differences between Archaeopteryx and Triceratops than between Archaeopteryx and Confuciusornis.

    I guess I just want to know where that inconsistency lies.

    1. Zach,

      You and I just do not see things in the same light. I myself can find no room in palaeontology or in any other science for arbitrary classification. We need clear definitions. As for your question, I think that the matter is very simple. Confuciusornis is a bird, and Archaeopteryx is a feathered dinosaur. They belong to two separate orders. Triceratops and Archaeopteryx are both reptiles and both dinosaurs. They are in the same order. To give you an analogy: cynodonts such as Thrinaxodon and Oligokyphus certainly bore more resemblance to mammals than to reptiles; but they still belong to the latter group, not the former, because they have not yet quite crossed over that evolutionary borderline that you cannot seem to perceive. Triceratops may look nothing like Archaeopteryx on the outside but he is a genuine reptile and so was Archaeopteryx. My final word for now: if the brainy dinosaur Troodon had ever evolved into a humanoid type, genetically and physiologically distinct from his ancestral trunk, then I would have been delighted to accept him as the founder of a new order. I think most palaeontologists would have done the same. Why can’t we accept the fact that there is a specific moment in time when a derived paravian’s chick hatched as the first full-blooded bird? Why can’t the serious science of palaeontology try a bit harder to identify that moment? Saying “It’s impossible to know when” is not an attitude that can encourage or foster further research.

  24. Mary, It’s possible you and I might have some disagreements about evolutionary processes — we’d have to talk more before I could be sure — but I do think that a great divergence from an evolutionary group that otherwise continues its own direction can be reason enough to name a new group, even at the same rank as the ancestral group. Why? Because classification is a practical process of dividing up biodiversity into useful groups (for learning, organizing information etc.). It should exist along side phylogenies — reconstructions of evolutionary history — and not as an alternative to them. All in all, arguing about which classification system to use is less helpful than learning more about the organisms, though sometimes fun.

  25. Words of gold, Barbara! In essence, I agree with you on this point. I would just respectfully ask the followers of the cladistic school to be more precise in their terminology and method of classification.

  26. By the way, the Linnaean classification system for plants was so arbitrary, so artificial, that it was abandoned fairly early for a system that put related plants together, even if they, for example, had different numbers of stigmas.

  27. Mary, Actually, the cladists are being precise in their classification and terminology, though sometimes they’ve changed the meaning of words, e.g. monophyletic, so early on and so forcefully that they’ve forgotten that they’ve made a change. (I hate to defend cladists, but unfortunately they’re not wrong though they are a bit over enthusiastic.)

    1. Barbara,

      Yes, I know that Linnean classification is not perfect (I’ve already admitted that in a previous post). But I do believe that it’s still the best we have at this time.

      In spite of what you’ve said, I do not see much precision in cladistics, just a lot of personal opinions.

  28. Barbara- “Despite the fuzziness of where birds start, there is a sense in which birds are different — we can call them birds rather than dinosaurs. And there’s a sense in which they are just really odd dinosaurs.”

    I guess it’s still arbitrary, but I think of ‘dinosaur’ as being roughly equivalent to ‘mammal’. There’s a lot of variation among mammals, but we still group them all together because of their shared traits. If flying mammals, bats, and entirely aquatic mammals, cetaceans, still get called mammals, why wouldn’t we still classify flying dinosaurs as dinosaurs? Sure, give them a new subgroup called birds, but still include them in the big group called dinosaurs.

    Granted, I’m not a scientist, but when I look at photos of an eagle and reconstructions of a deinonychus, and then compare that to photos of a dolphin and a horse, it seems that if any of those are going to be lumped into the same group, it would be the eagle and deinonychus.

    Like you said though, it’s an arbitrary distiction, and doesn’t change reality.

  29. Mary,

    I’m not trying to pick on your view, but some of it does seem confusing. For example, you wrote this:

    “I’m not claiming that brand new species arise brusquely from fundamentally different parents. Birds did not appear all at once, magically, from non-avian dinosaurs or even from early, non-derived “paravian” theropods. The development was gradual…”

    But then later, you wrote this:

    “Why can’t we accept the fact that there is a specific moment in time when a derived paravian’s chick hatched as the first full-blooded bird?”

    The first quote makes it seem like the normal gradual pace of evolution, but the second quote sounds more like saltationism. Am I misunderstanding something?

    Anyway, in regards to your second quote, I would say that the reason it can’t be pinned to a specific moment is because it’s not a specific moment. Let’s just pretend that an individual chick did hatch as the first full-blooded bird. Is it still the same species as its parents? If so, how can it be a bird, and its parents not birds? Who will it mate with? If it had that single last mutation to push it over into a true bird, but mated with an organism that was still not quite yet a bird, would half their offspring be birds and half not-quite birds? If you got back in a time machine and observed these animals, would there be any meaningful way to distinguish the true birds from the not-quite birds?

    I hope those examples make it clear why you can’t be so precise as to pick a specific moment in time when the first bird hatched. Biology just doesn’t work that way.

  30. Also, uh…mammals didn’t evolve from reptiles. They evolved from basal cynodont therapsids. The most basic split at the base of Amniota is between Sauropsida and Synapsida. The descendants of the former includes what we’d call “reptiles.” The descendants of the latter include mammals.

    So one didn’t evolve from the other. They’re separate groups, with separate evolutionary histories. The old term “mammal-like reptile” is a misnomer.

  31. Mary, you still haven’t addressed the single greatest flaw in your reasoning – why do you appear to accept that bats and whales are still mammals but claim that birds are no longer dinosaurs?

    To put it another way – Velociraptor and Triceratops are both dinosaurs. Velociraptor is much more closely related to birds than to Triceratops so how can birds not also be members of this same group of animals? That would be like saying that Ichthyornis and Pelecanus are one group (birds) but Passer is not.

  32. I’ve had similar discussions w/ppl who argued that birds aren’t dinos b/c birds are different. However, it was harder then b/c I didn’t know as much as I do now. I’ve since thought of what may be a quick & easy counter-argument (I say “may be” b/c I wanna get the opinions of others 1st): Just b/c you’re different from your ancestors doesn’t mean they’re no longer your ancestors; Saying otherwise is like saying that b/c I’m different from every other member of the Diaz family, I’m no longer a Diaz.

  33. I’m sorry that you consider cladistics, developed and improved during 25 years of peer-reviewed published research and accepted by the vast majority of workers to just be “personal opinions”. This is in contrast with your personal opinion that birds are not dinosaurs because you said so. (If you have any evidence to back up your claim, particularly if you’ve published your ideas in a peer-reviewed journal somewhere, I’m sure we’d all be glad to hear it).

    The truth is that the Linnaean system of classification was shown to be merely adequate for describing extant living things; eg birds are sufficiently different from mammals to be put in their own group. However, when you include extinct life, the Linnaean system fails to be of much use.

    This is why it has been abandoned by almost all palaeontologists in favour of cladistics. Dinosaurs have been shown by hundreds (if not thousands) of studies to be more closely related to birds than to lizards and turtles. Given that birds nest firmly within one of the two main groupings of dinosaurs – saurischia – this means that birds *must* also be dinosaurs (and, indeed, maniraptoran theropods). To deny this would either be illogical or indicate a wilful misunderstanding of relationships.

  34. Jeff,

    With regard to your latest comments, there is no contradiction in what I stated. The evolutionary process IS gradual, producing slow but meaningful modifications throughout the generations. These modifications eventually result in the birth of creatures which have effectively passed over the borderline of transition. And this crossover, this passage from one class or order to another, must necessarily occur at a specific moment.

    Let us consider the parents of the first bird (which may well have been Confuciusornis or Iberomesornis). These parents would have been borderline paravians, immediate proavians we might call them, in appearance perhaps almost indistinguishable from true birds. They themselves were the products of lengthy but progressive evolution. One momentous day, the female laid a batch of eggs. Two ore more of the offspring from this batch had structural and genetic differences which made them distinct from their transitional parents. And so, after a long process of evolution, IN A PARTICULAR MOMENT IN PREHISTORY, the first avian creatures made their appearance. (As two genetically compatible animals, male and female, are necessary to reproduce their own kind, I would have to say that whenever a crossover from one order to another takes place, there must be at least a pair of mutant offspring born of the transitional parents. This pair, male and female, would be responsible for the perpetuation of the new race.)

    To continue along the question of breeding, I have long pondered on this theme. It seems to me that, the further an evolving species or class deviates from its ancestral trunk, the less likelihood there is of mating with “its own kind”. The new, modified descendants would probably have had to practice inbreeding in order to preserve undiluted the acquired derived traits. Once over the borderline, once completely evolved into a new order, I affirm that there would no longer be any possibility of mating with their non-modified ancestors, toward the production of offspring. This would have become genetically impossible, just as Homo Sapiens cannot mate with monkeys and apes.

  35. Mark,

    You may see flaws in my argument, but I see only the clarity of definition. Bats and whales remain in Mammalia because they have the characteristic, recognized traits of mammals. Birds are no longer dinosaurs because they have lost too many of the structural and genetic traits of dinosaurs to be considered as such. Sure, some dinos like the maniraptors resembled birds. But, even if they were profusely feathered, they did not YET possess the full list of formal characteristics assigned to birds.

  36. Herman,

    Every living being born into this world is a unique individual. You, Herman Diaz, are different from your ancestors insofar as you have your own body, your own particular personality, your own destiny. But GENETICALLY speaking, you belong to and will always belong to the Diaz family. Your DNA ensures this.

    However, let’s say that, over many generations, there are physiological modifications in your descendants which result in the appearance of an extremely highly evolved, new type of being distinct from Homo Sapiens and superior in all aspects to him. This being would certainly have derived from the Diaz family, but would now belong to a different order, better and more highly suited for survival than man himself.

  37. It’s not just because I say so, Mark! Please understand that I am not expressing unfounded ideas pulled out of my own imagination. (I did NOT say, by the way, that cladistics represents ONLY personal opinion; I said A LOT of personal opinion).
    My lengthy, albeit independent, research has brought me into close contact with the studies of countless palaeontologists. I have printed out hundreds, maybe thousands of pages from scholarly tomes, journals, essays, articles and reports concerning the latest discoveries and research in avian evolution. Among the professional authors I have intensively consulted are Dr. Xu Xing and Prof.Luis Chiappe, who are certainly among the world’s foremost authorities in this branch of palaeontology. In sharing their ideas and classifications, do you think I am expressing merely personal opinions?
    When Aurornis was discovered, I conducted deep, careful research to determine whether this creature could have been a bird at all, not to say the first bird. After considering all available evidence, examining anatomical traits etc., I reached my own conclusions, yes, but based mostly upon what I had learned from the valid work of others.

  38. Andrea,

    You recently spoke of the “evolutionary continuum” and “intermediary forms”. Evolution throughout prehistory WAS continuous, always in flux, producing myriads of intermediary forms. But these modified life forms represented creatures which had diverged in some important way from their immediate ancestors, and more drastically separated from their remote ancestors. It’s a long, long way from the amoeba to Homo Sapiens. Well, everything may be ULTIMATELY derived from the first one-celled organisms. But so much beautiful, unique diversity has sprung up along the way. Continuity rolls along for a while until new, usually improved forms appear. I think that lumping special families and classes into super-orders only serves to blur the particular individual distinctions and cancels part of the marvelous uniqueness of each form of life.

  39. Mary, I appreciate the work you’ve put into this — but your comments continue to show that you misunderstand the fundamental issue you’re talking about here. The idea that all of our birds came from 2 (or more) individuals within one nest, because they’d have to have mates, simply isn’t consistent with what we know of evolution.

    I’ll also point out that your choice of “characteristic, recognized traits” is in itself arbitrary — if we chose a different set of traits, we could easily decide to separate out whales, bats, etc. as different orders. (Claiming that they’re no longer mammals, but something else…) The cladistic terminology attempts to explicitly use ancestry for classification, rather than forcing us to pick a particular set of traits.

    Tell me what traits make something a bird, rather than a dinosaur — and then try to use the same reasoning to construct a group that includes Triceratops and Compsognathus, but excludes Archaeopteryx. It’s going to have to involve a purely arbitrary choice of traits, since Compsognathus is much more closely related to Archaeopteryx than either is to Triceratops.

    1. psweet, I misunderstand nothing. You and I obviously just have a fundamentally different concept or interpretation of this fundamental issue. Is your viewpoint the only valid or plausible one in circulation? What do “we” know about evolution, if cladistics cannot even provide a precise definition of what a bird is? Several rather traditional palaeontologists have recently attempted to catalogue the characteristics which are proper to birds. I will have to look through my mountainous piles of photocopies, which are now inside crates following a move, but when I find these lists I will send you the links.
      “Characteristic traits” is not an arbitrary term. How could it be, when the fossil remains speak for themselves? Palaeontology has made huge
      strides in identifying and interpreting subtle details of anatomy; nowdays, impressions of skin, fur, feathers, scales,even sometimes internal organs can be observed with clarity; often the contents of the creatures’ last meal can be analyzed, unborn fetuses studied, pathologies determined. What is so arbitrary in classifying life forms according to the basic traits shared by all members of an order? The fact that Compsognathus is much more
      closely related to Archaeopteryx than either is to Triceratops does not remove any of the aforementioned animals from Reptilia.
      What makes a bird a bird? This is one of those tough-nut enigmas that has been stimulating palaeontological research and debate for decades. The mystery has not yet been resolved to everyone’s satisfaction…By no means! After having done considerable study of this phenomenon, I would, in broad
      terms, say that among the essential traits separating true birds from their theropod ancestors would be a beak, a reduced tail (in the form of a pygostyle), considerably prolonged feathered forearms in the form of wings which are no longer suitable for picking up objects or grasping, almost total loss of scaly tissue except on the legs, a strong markedly pronounced sternum conceived for “aerodynamic” flight…I could name other, more technical peculiar traits. Now, please don’t tell me that some of the derived paravians also had these characteristics…Some of them were evolving in that direction, yes, but they simply never achieved that high level of specialization which birds did.

  40. Post script to the previous message: in speaking of prehistoric birds here, I am referring to those which were capable of flight. They would all have had the characteristics mentioned above. At a later time, from these flighted birds others evolved, which were either adapted to aquatic environments (and may have ceased to fly due to their new watery habitat) or which
    perhaps because of corporal weight and large size had lost their ability to fly. The wings of these latter are usually quite reduced, almost to the point of seeming atrophied.

  41. In any case, whether one follows cladistic-based nomenclature or old-school nomenclature, taxa names are STATIC labels created arbitrarily by humans to describe the DYNAMIC process of evolution. If we had a complete fossil record preserving every single organism that lived, ir would be simply impossible to delineate taxa in an objective way. Therefore, IMO the lengthy discussions above are mostly useless.

  42. Ah, I see the problem. We’ve all been discussing science but Mary’s position is a philosophical one. It’s almost as if birds not being dinosaurs was the already assumed destination and then Mary has spent all this time looking for evidence to support that particular viewpoint. I believe that’s referred to as confirmation bias.

    When we do the science, we see that birds are actually not that different from some dinosaurs. In fact, those dinosaurs are more closely related to birds than they are to some other dinosaurs. Perhaps it’ll be clearer to use a crude diagram.

    [T(V I)]

    Here ‘T’ is Triceratops, ‘V’ is Velociraptor, and ‘I’ is Ichthyornis. Ichthyornis is a bird. Velociraptor is not a bird but is closely related to Ichthyornis. This is a natural grouping and we could choose to name this group if we thought that would be useful. Let’s call it Eumaniraptora.

    Now, Velociraptor and Triceratops are also in a natural group – the dinosaurs. If one member of Eumaniraptora is a dinosaur then so must be all members of Eumaniraptora. This is a simple mathematical relationship and to deny it would be to deny logic (although I’ve seen people who hold fast to a particular belief do this on many occasions).

    Lastly, the idea that all birds arose from an incestuous coupling of two nest-mates, the first of a New Order (sorry, Class), and somehow managed to not only survive, but flourish after generations of inbreeding contravenes all that we understand about breeding in animals and evolution in general. There is no evidence whatsoever that this ever happened and it has simply been made up to support the previously adopted position of birds not being dinosaurs.

  43. Mark,

    My last words absolutely on this forum, because any attempt to debate fruitfully here, without encountering self-contradiction, is practically impossible: you also are constantly looking for evidence to support YOUR particular viewpoint, and repeatedly present us with opinions about certain dubious “natural groups”.

    I think it’s little enough that cladistics has taught us about evolution. It’s merely one more interpretation among many. What’s so illogical about birds representing an Order? Do you honestly think that evolution is so static as to maintain the maniraptors indefinitely, as maniraptors, not as specialized “improved models”? It’s you who have started out from a premise which you wish to defend even against SCIENTIFIC logic. You speak of “incestuous coupling”; well, it must have been the same with “Adam and Eve”, the first two human beings. What else could they have breeded with? They managed not only to survive, but to reproduce and to flourish. You say there is no evidence whatsoever that this inbreeding ever happened. Can you provide me with evidence that it did not?

  44. Mary,

    Here’s a plain English summary of a study that did exactly that:

    Using genome sequences, the researchers reconstructed a history of effective human population size going back to the point when the chimp/bonobo and human lineages diverged. The tightest bottleneck was around 1000 individuals (note that the real population size was probably higher than the effective size). Most of the time, the population was far higher. And this includes all the different species between our last common ancestor with chimps and now.

    I’m not sure if you misunderstand the mainstream view of how evolution works, or simply don’t agree with it, especially speciation. Your comment about a new species occurring in a single generation is reminiscent of the hopeful monster of saltationism. At any rate, here’s a summary of the mainstream view of how speciation occurs within populations, not individuals:

  45. Final note:

    I was not thinking of posting here again, but I really must explain an inadvertent slip or imprecision I have made several times in this forum. Otherwise you might classify me as a “bird-brain”.

    I type fast and furiously, usually not re-reading what I’ve just written. This, I
    know, might be equated with carelessness, and it is a bad habit I wish to correct, as sometimes I find, too late, that in my haste I have typed embarassing errors. In strenuously stressing my point that birds are not dinosaurs, that they are outside of the (super)order dinosauria, my mind has dwelled on the word Order…What I have wished to say is that birds are beyond both Dinosauria and Reptilia, beyond a (super)order as well as
    beyond a class. Now, as the word “class” itself is currently being employed in more than one sense in palaeontological terminology, perhaps subconsciously I have avoided using it too much…But I do continue to claim that birds constitute a separate group, which is Class Aves, insofar as they are no longer either dinosaurs or reptiles. I certainly never meant to indicate that birds belong to a sub-order or group such as maniraptors.To speak in precise terms of
    taxonomy, birds are not therefore members of a new Order or sub-order, according to my argument, although in the more generic sense of “category” we may claim that they can be placed in a distinct order of fauna.
    My apologies for the occasional lack of precision. I am perfectly familiar with taxonomical terminology, and it was never my intention to misuse this or to confuse meanings.

  46. I tried leaving a comment earlier, but it seems to be held up in moderation, presumably because it contains a couple links. I’ll try again, breaking this up into two comments to try to get around the site’s spam filters. If the original ends up being approved and this becomes redundant, I apologize.


    Here’s a plain English summary of a study that did exactly what you requested regarding evidence about humans not originating from only two people:

    Using genome sequences, the researchers reconstructed a history of effective human population size going back to the point when the chimp/bonobo and human lineages diverged. The tightest bottleneck was around 1000 individuals (note that the real population size was probably higher than the effective size). Most of the time, the population was far higher. And this includes all the different species between our last common ancestor with chimps and now.

  47. Again to Mary,

    I’m not sure if you misunderstand the mainstream view of how evolution works, or simply don’t agree with it, especially speciation. Your comment about a new species occurring in a single generation is reminiscent of the hopeful monster of saltationism. At any rate, here’s a summary of the mainstream view of how speciation occurs within populations, not individuals:

  48. Mary, no matter how much birds change, no matter how “advanced” you think they have become, they will always remain dinosaurs, just as you will always remain a member of your family no matter how much you think you may have “changed”.

    Because clades are familial definitions.

    Reptiles, on the other hand, is not a clade, but a grade, which is characteristic based. It is perfectly viable to say that birds are no longer reptiles.

    However, there is no honest or rational or unbiased set of characteristics that could be used to define the grade of “reptile” in a way that excludes birds that would not also exclude all the maniraptoran dinosaurs, and likely the ceolurosaurs up to and including the tyrannosaurs as well.

    If you insist that birds are not reptiles, and remain honest and non-arbitrary in your definitions, then you must ALSO accept that some dinosaurs are similarly not reptiles, while other dinosaurs ARE reptiles.

    This is what inevitably happens if one insists on mixing clade and grade classification systems.

  49. Drat; just wandered in, late to the party as usual. Seems to me that Mary’s fundamental difficulty with the whole thing centers around definitions, as she said. The problem is, as far as my perhaps limited understanding goes, that almost all of our methods of classification involve some level of arbitrary distinctions, even cladistics (although cladistics strives to be ‘self-correcting’ in its attempts to find the ‘best’ organizational table). Even our definitions may be founded on somewhat arbitrary distinctions, and all too often we use terms without making clear what each of us understands those terms to mean. I suspect, given Mary’s method of expressing herself, that she is a disciple of the Olson/Feduccia school of bird evolution and classification. Having followed this with some interest for a few years now, it has seemed to me that at least over the last couple decades, Dr. Feduccia and his associates have drifted into what I generally call Procrustean Science – forcing the data to conform to their desired conclusion, and discarding those bits which don’t fit.. In this case, the desired conclusion is that Birds Are Not Dinosaurs, Nohow, No Way. I sympathize with the people involved, but find myself rather unsettled by the increasingly fanatic Certainty of the proponents. Science, I thought when I went in for it, is not another religion. Alas, people are only people.

Leave a Reply

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