ADVERTISEMENT
  About the SA Blog Network













Tetrapod Zoology

Tetrapod Zoology


Amphibians, reptiles, birds and mammals - living and extinct
Tetrapod Zoology Home

Dyke & Kaiser’s Living Dinosaurs: the Evolutionary History of Modern Birds

The views expressed are those of the author and are not necessarily those of Scientific American.


Email   PrintPrint



There are surprisingly few good books on the evolution and fossil history of birds: among those I recommend are Luis Chiappe’s Glorified Dinosaurs: The Origin and Early Evolution of Birds (Chiappe 2007), Gary Kaiser’s The Inner Bird: Anatomy and Evolution (Kaiser 2007), and Gerald Mayr’s Paleogene Fossil Birds (Mayr 2009). In view of this, Gareth Dyke and Gary Kaiser’s multi-authored Living Dinosaurs: the Evolutionary History of Modern Birds (published in 2011) is a most welcome addition.

An attractive volume with high production values and numerous excellent diagrams and photos (including a colour plate section), Living Dinosaurs contains 16 separate contributions on bird evolution, ranging in topic from bird origins and their Mesozoic and early Cenozoic diversification to conservation and the role of climate change in shaping the diversity and distribution of birds in the future. It is a technical book, intended for specialist researchers and not for a general audience.

As a strong proponent of the well supported inclusion of birds within the theropod dinosaur radiation, I’m personally more than happy to see people referring to birds as ‘living dinosaurs’. However, in this particular case I feel that the volume’s title is misleading, since it creates the impression that the book focuses on bird origins and on their place within Dinosauria more than it does. The volume is not, in fact, devoted to the evolutionary transition between non-avialan dinosaurs and birds, nor to the diversity and evolution of Mesozoic birds; just three of the volume’s articles cover these issues. Rather, it’s a very well rounded compilation of articles that cover the whole of bird history, the majority of included papers being on Cenozoic fossil birds and modern ones. Hopefully then, the title will not discourage those interested in birds but not necessarily in other dinosaur groups.

This reconstruction does not depict a bird - this is Anchiornis, a member of the bird sister-group, Deinonychosauria. During the Jurassic and Cretaceous there were several lineages of small, feathered maniraptoran theropods - birds were not 'special' or especially 'different' at this time. Reconstruction by Matt Martyniuk, licensed under Creative Commons Attribution 3.0 Unported license.

The 16 chapters are grouped into four sections. The first (containing three chapters) is on Mesozoic birds and on the evolution of  birds from among bird-like theropods; the last (containing a single chapter) is on conservation and climate change. However, the central two sections are not so obviously distinct and I’m hard pressed to see why some chapters are in one section (‘“The contribution of paleontology to ornithology”: the diversity of modern birds: fossils and the avian tree of life’) rather than the other (‘The evolution of key avian attributes’).

Mesozoic birds and other dinosaurs

A selection of Mesozoic birds. Top to bottom: skeletal reconstruction of Sapeornis (by Darren Naish, redrawn from Zhou & Zhang 2003); life reconstruction of enantiornithine Longipteryx (by Nobu Tamura); and skeleton of Patagopteryx (by FunkMonk). Longipteryx and Patagopteryx images licensed under Creative Commons Attribution 3.0 Unported license.

Among the chapters in that first section, Makovicky and Zanno’s (‘Theropod diversity and the refinement of avian characteristics’) and O’Connor, Chiappe and Bell’s (‘Pre-modern birds: avian divergences in the Mesozoic’) are both strong, providing good reviews of Mesozoic bird history. O’Connor et al.’s lengthy review of Mesozoic bird taxa is thorough and extremely well illustrated; the chapter is further useful in providing a large cladistic analysis. Unfortunately, there is hardly any resolution within Enantiornithines but a more recent study that builds on the same dataset does depict far better resolution within this large and important clade (O’Connor & Zhou 2012).

The third chapter in this section – Ward and Berner’s ‘Why were there dinosaurs? Why are there birds?’ – discusses how atmospheric composition might have contributed to Mesozoic dinosaur evolution. An interesting area no doubt, especially since improved knowledge of dinosaurian pneumaticity and the Mesozoic atmosphere have both inspired many to ask questions about the same topic. Ward and Berner suggest that dinosaur diversity may have increased in step with rising oxygen levels and that apneumatic ornithischians required these higher levels in order to flourish.

However, it looks unlikely that their data on dinosaur diversity across the Mesozoic is fine-scale enough to reliably support these suggestions. Furthermore, the assumption that dinosaur diversity was driven by such factors as atmospheric composition ignores the fact that dinosaurs were biologically remarkable, with (among saurischians) a bird-like pneumatic system, a predisposition for the evolution of long, flexible necks, an often quadrupedal, parasagittal construction that encouraged the evolution and diversification of large-bodied forms, and a life cycle that – even among giant forms – involved rapid growth and the production of relatively large egg clutches. Indeed, empirical studies have found little support for any link between dinosaur diversification and atmospheric composition: instead, the diversification potential (and hence general biology) of dinosaurs themselves was the main factor driving their evolution (Sander et al. 2011, Sookias et al. 2012).

Livezey, pseudotoothed birds and phorusrhacids

Many will note that the book includes one of the last (if not the last) published contributions submitted by Brad Livezey (sadly, Brad died in a vehicle collision in February 2011). His article is a philosophical review of the problems and pitfalls facing avian phylogenetics. There is much criticism here of the approaches favoured by various workers across palaeornithological and molecular fields, and there’s little doubt that some or many of his assertions will irk colleagues. It is certainly interesting to see Livezey’s opinions on such topics as the proposed (and, actually, robustly supported) close relationship between grebes and flamingos, and also his take on such issues as the ‘marriage of convenience’ between “paleomorphological and neomolecular” datasets (Gerald Mayr’s work is cited heavily in this section of the text), phylogenetic parsimony vs realism, and supertrees.

Life reconstruction of the Paleogene pseudotoothed bird Dasornis toliapica, by Estelle Bourdon. This image appears as Plate 8.6 in Dyke & Kaiser (2011). (c) E. Bourdon.

Several chapters review the current state of knowledge of key Cenozoic fossil groups, making the book an essential reference for palaeornithologists. Ksepka and Ando’s review of penguin evolution and history, and Bourdon’s chapter on pseudotoothed birds, are strong and valuable contributions (Bourdon’s suggestion that pseudotoothed birds are close to anseriforms has since been evaluated by Mayr (2011). He finds pseudotoothed birds to be the sister-group to the whole of Galloanserae).

Alvarenga, Chiappe and Bertelli review phorusrhacids and provide a new phylogenetic analysis of the group. Bathornis and Elaphrocnemus are recovered as successively closer to Phorusrhacidae (both are closer than is Cariama), and Phorusrhacidae includes Mesembrionithinae, Psilopterinae, and a mostly unresolved clade that includes patagornithines, phorusrhacines and brontornithines (Alvarenga et al. 2010). For all their fame as fantastic animals and among the most awesome of birds ever, phorusrhacids have seemingly been rather neglected as objects of serious study until quite recently. When we combine this chapter with Alvarenga & Höfling’s (2003) systematic revision of the group and Degrange et al.’s (2010) analysis of skull mechanics, we have an exciting new body of work on these birds, sure to inspire future studies. [There’s a lot of material on phorusrhacids in the Tet Zoo archives. I should collate, update and republish it all some time. See links below.]

Reconstructions (by Eduardo Brettas) depicting phorusrhacid diversity (and one seriema), from Alvarenga et al. (2011). A, Cariama. B, Mesembrionis. C, Psilopterus. D, Andalgalornis, E, Phorusrhacus. F, Paraphysornis, G, Brontonis. The human figure is 1.75 m tall. (c) E. Brettas.

Molecular (rather than fossil) data has been key to the construction of a modern phylogeny for the astonishingly species-rich passerines; Barker’s chapter reviews recent developments in this area, covering the structure of the tree as a whole, biogeography, and the whole question of why passerines (and why certain lineages within the group itself) are so diverse.

WAIR, brains, and the future

Tobalske et al.'s (2011) images, showing (1) airflow around chukar chick practising WAIR, and flow field (2a) around hummingbird during downstroke and (2b) upstroke. Images originally plates 10.6 and 10.13 in Dyke & Kaiser (2011).

Part 3, on key avian attributes, includes chapters on the evolution of the flight stroke, brain anatomy, the timing of avian evolution, and the functional and phylogenetic diversity of marine birds. Tobalske, Warrick, Jackson and Dial’s article on the ‘Morphological and behavioural correlates of flapping flight’ combines another take on wing-assisted incline running with discussions of intermittent flight (bounding flight, flap-bounding and so on), manoeuvrability, and hovering (lots on hummingbirds, of course). Walsh and Milner’s article on avian brain anatomy and senses is an excellent review of this subject, even including a discussion of reported fossil endocasts (and claimed endocasts: witness Cerebavis cenomanica from the Late Cretaceous of Russia). I also especially enjoyed Kaiser’s discussion of those details of avian anatomy linked to aquatic life, but then I am something of a fan of Kaiser’s work (Kaiser 2007, Naish 2011).

Part 4 is titled ‘The future: conservation and climate change’. It only includes a single article – Thomas’s ‘The state of the world’s birds and the future of avian diversity’ – but it is a particularly good one, focusing on the predisposition of avian lineages to threat, the ecological consequences of their decline, and the responses of bird lineages to climate change. We know that some birds are seemingly benefiting from climate change (one example: Wandering albatrosses Diomedea exulans grow faster, get bigger and breed more successfully at the moment due to faster winds in the Southern Hemisphere (Weimerskirch et al. 2012)), but studies generally indicate that bird distribution and breeding behaviour is not changing fast enough to keep track with climate change. The article is a great review of the work done so far; it’s an area of research that’s sure to become more important in coming decades.

In short, Living Dinosaurs is a most worthy and well crafted volume. Its strength is in providing a surprising number of really good reviews of many aspects of bird evolution and history, generally written by leading workers in the respective areas. I personally found the book highly useful in my own research and ended up citing many of its chapters in a recently published review of the avialan fossil record (Naish 2012).

Gareth Dyke & Gary Kaiser, 2011. Living Dinosaurs: the Evolutionary History of Modern Birds. John Wiley & Sons (Chichester, UK), pp. 422. ISBN 978-0-4706-5666-2. Hardback, index, refs. Here on amazon. Here on amazon.co.uk.

For previous articles on some of the subjects mentioned or discussed here, see…

Refs – -

Alvarenga, H., Chiappe, L. & Bertelli, S. 2011. Phorusrhacids: the terror birds. In Dyke, G. & Kaiser, G. (eds) Living Dinosaurs: the Evolutionary History of Modern Birds. John Wiley & Sons (Chichester, UK), pp. 187-208.

Chiappe, L. M. 2007. Glorified Dinosaurs: The Origin and Early Evolution of Birds. John Wiley and Sons, Hoboken (NJ, USA).

Degrange, F. J., Tambussi, C. P., Moreno, K., Witmer, L. M. & Wroe, S. 2010. Mechanical analysis of feeding behavior in the extinct “terror bird” Andalgalornis steulleti (Gruiformes: Phorusrhacidae). PLoS ONE 5 (8): e11856. doi:10.1371/journal.pone.0011856

Kaiser, G. W. 2007. The Inner Bird: Anatomy and Evolution. University of British Columbia, Vancouver.

Mayr, G. 2009. Paleogene Fossil Birds. Berlin, Springer.

Mayr, G. 2011. Cenozoic mystery birds – on the phylogenetic affinities of bony-toothed birds (Pelagornithidae). Zoologica Scripta 40, 448-467.

Naish, D. 2011. [Review of] The inner bird: anatomy and evolution. Historical Biology 23, 313-316.

Naish, D. 2012. Birds. In Brett-Surman, M. K., Holtz, T. R. & Farlow, J. O. (eds) The Complete Dinosaur (Second Edition). Indiana University Press (Bloomington & Indianapolis), pp. 379-423.

O’Connor, J. K. & Zhou, Z. 2012. A redescription of Chaoyangia beishanensis (Aves) and a comprehensive phylogeny of Mesozoic birds. Journal of Systematic Palaeontology DOI:10.1080/14772019.2012.690455

Sander, P. M., Christian, A., Clauss, M., Fechner, R., Gee, C. T., Griebeler, E. M., Gunga, H.-C., Hummel, J., Mallison, H., Perry, S., Preuschoft, H. Rauhut, O., Remes, K., Tütken, T. Wings, O. & Witzel, U. 2011. Biology of the sauropod dinosaurs: the evolution of gigantism. Biological Reviews 86, 117-155.

Sookias, R. B., Benson, R. B. J. & Butler, R. J. 2012. Biology, not environment, drives major patterns in maximum tetrapod body size through time. Biology Letters 8, 674-677.

Weimerskirch, H., Louzao, M., de Grissac, S., & Delord, K. 2012. Changes in wind pattern alter albatross distribution and life-history traits. Science 335, 211-214.

Zhou, Z. & Zhang, F. 2003. Anatomy of the primitive bird Sapeornis chaoyangensis from the Early Cretaceous of Liaoning, China. Canadian Journal of Earth Sciences 40, 731–747.

Darren Naish About the Author: Darren Naish is a science writer, technical editor and palaeozoologist (affiliated with the University of Southampton, UK). He mostly works on Cretaceous dinosaurs and pterosaurs but has an avid interest in all things tetrapod. His publications can be downloaded at darrennaish.wordpress.com. He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at tetzoo.com! Follow on Twitter @TetZoo.

The views expressed are those of the author and are not necessarily those of Scientific American.





Rights & Permissions

Comments 60 Comments

Add Comment
  1. 1. gesimsek 12:49 pm 08/26/2012

    Is it possible to speculate that birds were the hot-blooded version of dinosaurs? Hence, their cold-blooded dinosaurs died when meteor hit the earth and changed the climate so that all cold-blooded animals died (except the ones living under ground) and the age of hot-blooded mammals started.

    Link to this
  2. 2. BrianL 1:02 pm 08/26/2012

    Looks like I have another book to order. I’m rather pleased that I already have two of the other titles you recommended! Speaking of which, I do find it rather bizarre that Paleogene fossil birds is, well, limited to discussion of Paleogene fossils. In some accounts, discussion of a group simply stops despite there sometimes being quite a bit to say about their Neogene representatives.

    I see that brontornithines are apparently included within phorusrhacids without ifs, buts or maybe’s. Is that the case? It would seem rather sloppy to let *Brontornis*’ reevaluation as a possible anseriform (relative) go without mention. Also, how realistic is that insanely bigheaded *Brontornis* illustration? It looks cartoony. How much do we know of *Brontornis*’ physical appearance anyway? Also, if galloanseraean, how likely is that it was a dedicated carnivore?

    That cariamids and phorusrhacids apparently reached South America independently is interesting though not terribly surprising. Still, Paleogene Cariamae already seem rather prone to being flightless yet they apparently were good dispersers. Given the size of their closest living relatives, this makes me wonder how small and volant ancestral cariamaeans were and how prone they were to fossilisation.

    That pelagornithid looks very much like an albatros with a big pseudotoothed beak glued to it. Is it an analogy taken too far or were pelagornithids truly that albatross-like? Fpr what it’s worth, I’ve always greatly appreciated the way they were illustrated in ‘The origin and evolution of birds’. Too bad about the text in that book…

    @gesimsek:
    It’s better not to talk about hotblooded and coldblooded as if they’re two non-overlapping and absolute terms. From what I gather, these days those terms are seen as outdated because they deny the great diversity in metabolisms in animals. I think the wikipedia article on animal metabolism are quite good, perhaps you should check them out.
    Nevertheless, birds (as in neornitheans) do have a very active metabolism but if it exceeds that of other dinosaurs, it’s probably better to see them as being on the extreme end of a continuous scale among dinosaurs than as being ‘metabolically opposed’ to other dinosaurs. Dinosaurs in general were pretty active animals, it seems.
    Note that quite a few mammals (monotremes and sloths come to mind) are not nearly as ‘hot-blooded’ as neornitheans are. In fact, birds are metabolically more efficient than mammals in general. I wouldn’t be surprised if many other dinosaurs had a far more active metabolism than, say, sloths.

    I’m moving on thin ice here though, given that physiology is neither my forte nor my principal interest as dinosaurs go.

    Link to this
  3. 3. Andreas Johansson 3:04 pm 08/26/2012

    phylogenetic parsimony vs realism

    Would you mind explaining what this is about, Darren?

    Link to this
  4. 4. JoseD 5:35 pm 08/26/2012

    @Darren Naish

    “There are surprisingly few good books on the evolution and fossil history of birds: among those I recommend are Luis Chiappe’s Glorified Dinosaurs: The Origin and Early Evolution of Birds (Chiappe 2007), Gary Kaiser’s The Inner Bird: Anatomy and Evolution (Kaiser 2007), and Gerald Mayr’s Paleogene Fossil Birds (Mayr 2009).”

    Don’t forget about “Dino-Birds”. It’s not as recent as the above books, but “The Natural History Museum Book of Dinosaurs” is & Chapter 10 is basically an updated version of “Dino-Birds”.

    “Walsh and Milner’s article on avian brain anatomy and senses is an excellent review of this subject, even including a discussion of reported fossil endocasts (and claimed endocasts: witness Cerebavis cenomanica from the Late Cretaceous of Russia).”

    That was my favorite chapter. I especially liked page 296-298.

    @gesimsek

    Actually non-avian dinos probably weren’t ectothermic (I.e. Cold-blooded). Non-avian maniraptors were probably endothermic (I.e. Hot-blooded) & non-maniraptoran dinos were probably mesothermic (I.e. In btwn ectothermic & endothermic): go here.

    Link to this
  5. 5. Tayo Bethel 2:16 am 08/27/2012

    I have this book–havent had a chance to sit down and read it through though. Now if I could only get my hands on The Inner Bird: Anatomy and Evolution I would be very very happy. LOL My list of books I simply must read is getting longer and longer and longer …

    Link to this
  6. 6. Perisoreus 2:49 am 08/27/2012

    Andreas Johansson:

    I think Darren means that in reality, trees are often enough not most parsimonious and that genomes are not as conservative when it comes to mutating as previously thought.

    Link to this
  7. 7. naishd 3:18 am 08/27/2012

    Thanks much for comments. Some responses…

    Re: could birds be regarded as an endothermic branch of the otherwise ectothermic Dinosauria (comment 1)?

    Essentially, no. Firstly, early birds (Archaeopteryx and so on) were almost certainly not really different from closely related, bird-like dinosaurs: modern birds (neornithines) may well have elevated metabolic rates relative to Mesozoic dinosaurs, but those metabolic rates are elevated compared to those of Mesozoic birds as well. Secondly, were Mesozoic dinosaurs ectothermic? Some people have assumed this because (1) non-avialan dinosaurs are “reptiles”, and (2) they see problems with claims that dinosaurs were endothermic, but I’m not aware of any good evidence that actually provides plausible support for the presence of ectothermy in these animals. Argument continues of course, but endothermy of some sort probably evolved in dinosaurs long before birds were ‘invented’.

    Darren

    Link to this
  8. 8. naishd 3:37 am 08/27/2012

    More responses…

    Re: Brontornithines being included among phorusrhacids (comment 2). “Alvarenga et al. (2011) have since argued that the characters used to support [Agnolin's inclusion of Brontornis within Anseriformes] are either erroneous or present in undoubted phorusrhacids, and that Brontornis can be confidently assigned to Phorusrhacidae” (Naish 2012, p. 407). Refs as above.

    Re: Bourdon’s reconstruction of Dasornis – yeah, other people have opined that these animals should be made to look more like pelicans than albatrosses. Mike Habib is the guy to ask – I know he’s done some work on wing loading, flight performance etc. in these birds. The length and form of their wing bones suggests that their wing shape may well have been albatross-like, but I’d love to hear more on this. The illustrations you’re referring to from Feduccia’s book, by the way, were done by Mark Hallett. It’s a shame they’re not available elsewhere in the literature.

    Darren

    Link to this
  9. 9. naishd 4:06 am 08/27/2012

    From comment 3…

    Re: phylogenetic parsimony vs realism

    “Would you mind explaining what this is about, Darren?”

    Livezey includes a two-page discussion (pp. 133-134) of ‘Parsimony: philosophical pedigree and simplicity vs realism’ – his argument is complex and it’s not at all easy to subtract a basic takehome from what he says. Here is some of what he says…

    “Whereas most applications of parsimony in phenotypic contexts have performed to the expectations of morphological practitioners, the criterion met with less acceptance among those analyzing DNA sequences. However, most in both camps employ criteria of parsimony in various ways, e.g., adoption of the least complex model of comparable explanatory power or the parsimony implicit in sequence alignment (Anderson, 2008). Still others defend the comparative explanatory power of parsimony (Farris, 2008).

    Some detractors of parsimony consider that it does not assume a probabilistic model of evolution or that the criterion is not intrinsically related to the evolutionary process, some consider parsimony equivalent to a model of “minimal evolution” (Rzhetsky & Nei, 1993; Penny et al., 1994; Gascuel, 2000; Steel & Penny, 2000; Denis & Gascuel, 2003), and still others condemn it by the claim that parsimony actually is a model of evolution but is burdened by an exorbitant number of implicit parameters (Felsenstein, 2004). Although for a given data-set the search for minimal steps or change is consistent with this view, the terminology incorrectly implies that those using the parsimony criterion are limited by a preconception that evolutionary rates and changes are rare, whereas the method only seeks a minimalist explanation for data sets characterized by any of a range of evolutionary rates. Morever, although parsimony favors topological solution at least as efficient (i.e. of minimal homoplast) in the explanation of characters among lineages (other parameters being equal), it is virtually routine to consider “suboptimal” or implications of longer topologies in such studies (Sober, 1983a)” (p. 133). And he says a lot more on the issue as well.

    Darren

    Link to this
  10. 10. Jerzy v. 3.0. 4:09 am 08/27/2012

    Re: climate change
    1. I read long ago, I think in one of Roger Tory Peterson’s popular books, that it is estimated that in Pliocene, with warmer climate, there were 20% more bird species than today. It stuck me as incredibly cool.

    I wonder if there is any serious analysis of this kind?

    2. Was this model about birds responding to climate change tested on Medieval warm period, Lower Dryas cooling and Eemian warming and replicated that all the living species survived?

    I think one problem is that models assume that birds should exactly follow the changing climate and any lag is bad. There is no reason to believe so. Birds follow habitats, and will change their distribution only if farming practices and forest composition will change to follow the climate. I guess forest birds will show a lag in response equal at least to the lifespan of forest trees, so a century or so.

    Link to this
  11. 11. naishd 4:24 am 08/27/2012

    Did pseudotoothed birds really resembles albatrosses in wing form? I asked Mike Habib, and he said…

    “Based upon the feather impressions of the California specimens, the planform appears to be albatross-like. Cross-sectional geometry of the limb elements supports this conclusion. The wing impressions are too distorted to get anything more than an estimate of Aspect Ratio. However, at high aspect ratios, tip slots lose effectiveness, and no living bird with an AR over 11.5 has slotted tips. Therefore, we can deduce that pelagornithid wings were of high AR and slot-less, which makes them effectively albatross like in overall morphology.”

    Thanks indeed to Mike for this useful response. I forgot that feather impressions are known for members of the group.

    Darren

    Link to this
  12. 12. BrianL 8:33 am 08/27/2012

    Thanks for the reaction. A pity *Brontornis* is a phorusrhacid after all then, for evolutionary reasons. Its status as a particularly extreme phorusrhacid is cool though. Still I wonder if anyone knows how realistic that grotesque reconstruction of *Brontornis* is.

    Speaking of mysterious, large and extinct South American birds, does the book include any mention of *Cladornis*? Or *Eremopezus*, for that matter?

    Link to this
  13. 13. Andreas Johansson 10:27 am 08/27/2012

    @Darren: Thanks.

    Link to this
  14. 14. naishd 3:51 pm 08/27/2012

    Forgot to say… I don’t think the phorusrhacid reconstructions shown above are particularly accurate – yes, they do look somewhat, well, cartoony.

    Darren

    Link to this
  15. 15. John Harshman 4:28 pm 08/27/2012

    There is now an ad for Scientific American completely covering the top half of this post, and the only way I can get rid of it is to “read more”. This, as well as the complete substitute window I often get when pulling up Tet zoo, is a very annoying hard sell. Is there any way to get them to stop?

    Link to this
  16. 16. Bill_Crofut 6:21 pm 08/27/2012

    Dr. Naish,

    Re “…dinosaurs were biologically remarkable, with (among saurischians) a bird-like pneumatic system…”

    How bird-like was the pneumatic system?

    For example:

    “Pneumatic bone — A hollow bone filled with air, connected to the respiratory system. Many of a bird’s bones are pneumatic, with struts across their hollow interiors to provide a combination of light weight and strength as an adaptation to flying.

    http://www.birdcare.com/bin/showdict?pneumatic+bone

    Was it that bird-like?

    [from Darren: Mr Crofut - maybe you should do some research yourself, instead of hoping to find weaknesses in statements made about the evolutionary relationship between birds and other dinosaurs. As explained in comments below: yes, the air-sac system of non-bird dinosaurs was essentially identical to that of modern birds. Since you are only here in an effort to poke holes and make trouble, this is your last comment. Anything further will be deleted. I assume you are the same Bill Crofut who argues elsewhere online that the sun goes round the Earth.]

    Link to this
  17. 17. THoltz 8:47 pm 08/27/2012

    Bill_Crofut: in the vertebrae: yes, that bird-like. In other bones: sometimes in some taxa.

    To all: one issue with the O’Connor et al. (wonderful) paper is the lack of an electronic version of the matrix. It would be wonderful to run this thing oneself, and find out if the irresolution of Enantiornithes is real lack of resolution, a fully-resolved single tree with a single nastyass rogue taxon, or (more likely) a pretty well-resolved tree with a few rogues.

    Link to this
  18. 18. THoltz 8:54 pm 08/27/2012

    Recent review of distribution of bird-like pneumatic structures in various archosaur groups.

    Link to this
  19. 19. David Marjanović 7:09 am 08/28/2012

    Aww, Darren, the banhammer ruins all the fun :-)

    BTW, Dasornis “toliapica” is of course toliapicus. Ornis is a he, not a she.

    Ward and Berner suggest that dinosaur diversity may have increased in step with rising oxygen levels

    I don’t understand why oxygen would lead to cladogenesis. I can imagine body size or even abundance increasing with oxygen levels, but why diversity?

    the terminology incorrectly implies that those using the parsimony criterion are limited by a preconception that evolutionary rates and changes are rare, whereas the method only seeks a minimalist explanation for data sets characterized by any of a range of evolutionary rates.

    In Sereno’s words, IIRC: parsimony doesn’t assume the minimization of convergence, it minimizes the assumption of convergence.

    Medieval warm period

    That wasn’t a global event. It happened in various patches of the northern hemisphere, and in different centuries in each.

    There is now an ad for Scientific American completely covering the top half of this post, and the only way I can get rid of it is to “read more”. This, as well as the complete substitute window I often get when pulling up Tet zoo, is a very annoying hard sell. Is there any way to get them to stop?

    I use Firefox with Adblock Plus; that’s probably why I’ve never had the problems you mention.

    Link to this
  20. 20. Dartian 7:34 am 08/28/2012

    Jerzy:
    I read long ago, I think in one of Roger Tory Peterson’s popular books, that it is estimated that in Pliocene, with warmer climate, there were 20% more bird species than today.

    That exact claim does not sound familiar to me, but I do recall that in James Fisher and Roger Tory Petersons’s World of Birds (originally published in 1964; II edition published in 1988) there is some brief speculation on the number of bird species that have existed during the Cenozoic. I don’t have the book at hand at the moment and I don’t remember precisely what the authors said, but I recall that Fisher’s personal belief was that the total number of bird species ever to have existed was more than half a million. (James Fisher had died before the first edition of WoB was published. No explanation was given for how he arrived at the 500,000+ figure, and no explanation was given for why Peterson, apparently, thought that number to be too high. Apparently these were just pure guesstimates.)

    I wonder if there is any serious analysis of this kind?

    I’d guess not; most likely, there simply aren’t enough data for that.

    Link to this
  21. 21. naishd 10:05 am 08/28/2012

    On our creationist friend (and David’s response in # 19), my impression by and large is that people dislike seeing creationists on the site – as Dartian said last time, they just waste time and cause needless diversions. I’m more interested in seeing other kinds of debates, so my policy is to ban. If enough people (genuine people, not sockpuppets!) object I will consider. Speak now or forever hold your peace.

    Darren

    Link to this
  22. 22. David Černý 10:07 am 08/28/2012

    @Dartian:

    John Boyd’s Taxonomy in Flux quotes one volume of HBW* as stating that the total number of bird species to have existed is close to 1.6 million. I don’t have the book in question (perhaps someone who does could confirm it?) so I don’t know how the number was calculated, but it seems that the assumption of higher Cenozoic bird diversity must have been involved to arrive at such a huge figure.

    *The full reference is:

    Caley KJ 2007 Fossil birds. 10–55 in del Hoyo J, Elliott A, Christie D, eds. Handbook of the Birds of the World, volume 12, Picathartes to Tits and Chickadees. Barcelona: Lynx Edicions

    Link to this
  23. 23. naishd 10:22 am 08/28/2012

    On how many birds existed in the past, Jerzy and Dartian are referring to this passage in Fisher and Tory Peterson’s 1988 World of Birds

    “A rough calculation by Professor Pierce Brodkorb suggests that rather over 1-and-a-half (actual text includes fraction) million bird species have existed at one time or another in the 140 million years that have elapsed since the first bird appeared. James Fisher’s own belief is that the number is rather under half a million” (p. 75).

    Brodkorb’s estimate is repeated in a few other places in the literature. It all comes from…

    Brodkorb, P. 1960. How many species of birds have existed? Bulletin of the Florida State Museum, Biological Sciences 5, 41-53.

    Which you can read here. Incidentally, Brodkorb seems to have been a thoroughly brilliant person with an excellent sense of humour and a passion for collections of bones and piles of literature. His 1993 obituary (written by Storrs Olson) is well worth a read.

    Darren

    Link to this
  24. 24. Jerzy v. 3.0. 10:50 am 08/28/2012

    Thanks folks, I must have mis-constructed something.

    About the total number of bird species ever existing, calculation depends from the estimated length of species existence. Tricky to estimate in birds, as small passerines have near-identical bones.

    About number of bird species in the past, such calculation might have some more scientific basis. There are correlations between temperature and species diversity, so one might estimate number of bird species on the whole Earth in the warmer period. Caveat – important would be sea rise, potentially splitting islands which would then develop endemic species!

    Link to this
  25. 25. Jerzy v. 3.0. 10:52 am 08/28/2012

    A… the issue deal with both! Going reading!

    Link to this
  26. 26. Heteromeles 11:05 am 08/28/2012

    @19: David, Peter Ward’s Out of Thin Air lays out his argument for how fluctuations in atmospheric [O2] could lead to cladogenesis. I loved the book when I first read it, but I notice that it’s sunk without a trace amongst the paleontological community. I don’t know if that’s because it’s fundamentally wrong or what.

    Anyway, the general idea is that episodes of high [O2] promote new clades, because during such times, animals can “afford” to have inefficient lungs. This might favor, among other things, aquatic animals becoming terrestrial, with inefficient lungs. In general, his idea is that high [O2] allows inefficiency, and thus radically new, but inefficient, body plans have a chance to survive. Periods of low [O2] do the opposite, favoring the animals with the most efficient lungs.

    His theory is that dinosaurs became dominant due to a massive decrease in [O2] during the Triassic, and they simply had more efficient lungs than did mammals, reptiles, or insects, all of which were limited to small bodies with high surface area:volume ratios just so they could breathe. He infers that mammals eventually took over after the K-T extinction in part because [O2] levels were high enough that mammals were no longer limited to small body sizes, either by [O2] or by competition with larger dinosaurs.

    The nice thing about Ward’s hypothesis (really hypotheses) is that it provides an explanation for why dinosaurs would have highly advanced lungs and tiny brains: low [O2]. Assuming his reconstruction of [O2] is accurate (increasing over the Mesozoic), he provides a hypothesis for things like increasing brain size over time and gigantomorphy in the Jurassic and Cretaceous, but not in the Triassic.

    The biggest problem I see with his book is that it’s not clear whether his reconstruction of [O2] through time is accurate. If it isn’t, then many of his ideas fall apart.

    Link to this
  27. 27. Dartian 11:17 am 08/28/2012

    Darren at #23:
    Oh, bugger! I thoroughly mis-remembered what Fisher & Peterson said. Oh well, that’ll teach me to try and cite anything without the actual reference in front of me…

    Link to this
  28. 28. naishd 11:23 am 08/28/2012

    On Ward’s stuff about O2 levels… it hasn’t won out because it probably isn’t accurate. The rising and failing fortunes of animal groups don’t reliably match with atmospheric composition: see Sander et al. (2011) and Sookias et al. (2012), both cited above. Both sets of authors failed to find a good match between body size evolution and atmospheric composition. Even among insects – perhaps the one group whose evolution and body size variation might most logically be linked with atmosphere – there are good reasons for thinking that biological factors (namely, the evolution of other groups of flying animals) were more important drivers than atmospheric composition.

    With all due respect to Ward and others who have looked to the primacy of atmosphere as a controlling factor in shaping evolution, I think people are too keen to look for magic, abiotic factors to explain the weird life of the past when the real answers lie in the remarkable anatomy and physiology of the groups concerned. Dinosaurs – to take the famous example – didn’t get bigger than mammals because dinosaurs were evolving in unusual gravitational, atmospheric or thermal conditions. Dinosaurs got bigger than mammals because they were fundamentally better at doing it. Or, if you like, dinosaurs were more awesome (smiley).

    Darren

    Link to this
  29. 29. Heteromeles 12:04 pm 08/28/2012

    @Darren: I don’t particularly want to get into a fight defending Ward, especially because (as I noted above) I wasn’t sure his atmospheric reconstructions were accurate to begin with.

    Still, I think it’s a bit much to say that things like sauropod pneumaticity are independent of the atmosphere. Certainly, by the time we’re talking about giant sauropods, they were. Unfortunately, one can justly argue that the reason that animals with air sacs were around to evolve to gigantism was because their ancestors were the most efficient and biggest things around during the formation of the Siberian traps.

    The problem with strong selection events is that they have long echoes. While I agree that dinosaurs were awesome, I can’t help but notice that last round of giant theropod wannabes are all extinct, unlike what happened during the Mesozoic after mass extinction events. I don’t mind considering [O2] as one of those baseline issues. For example, mammals can’t get as big as dinosaurs for a given [O2]. Still, if you start with small theropods and small mammals (as in the Danian Paleocene), mammals can evolve to outcompete theropods if their body size isn’t limited by low [O2]. I think the fossil record pretty clearly demonstrates this, if nothing else.

    Link to this
  30. 30. naishd 12:22 pm 08/28/2012

    “Unfortunately, one can justly argue that the reason that animals with air sacs were around to evolve to gigantism was because their ancestors were the most efficient and biggest things around during the formation of the Siberian traps.”

    You mean – - round about the Permo-Triassic boundary? As in, long before archosauromorphs had evolved pneumaticity?

    Darren

    Link to this
  31. 31. John Harshman 1:27 pm 08/28/2012

    Caley KJ 2007 Fossil birds. 10–55 in del Hoyo J, Elliott A, Christie D, eds. Handbook of the Birds of the World, volume 12, Picathartes to Tits and Chickadees. Barcelona: Lynx Edicions

    I have that. But the estimate of total bird species there is just taken from Brodkorb (1959 — where does the 1960 bit come from?), just like the others. Oddly, it’s very precise: 1,634,000. The number of wild-ass guesses in that estimate is very, very high; insufficient, I think, to have any confidence in the result, even within an order of magnitude. Brodkorb was also writing a long time ago, and many of his assumptions have been superseded, for example his notion that most Cretaceous birds were aquatic. We also get into arguments about species concepts, as there are a good thousand or more extant bird species than when he wrote, apparently.

    …assumptions of parsimony

    At the risk of starting a war, parsimony assumes that all branches are short (in terms of expected number of changes) and of equal expected length (or, alternatively, that the probability of change in one character tells you nothing about the probability of change in any other characters along the same branch). Of course, large enough violations of these assumptions will produce incorrect trees, as Felsenstein (1978) showed.

    Brodkorb, P. 1959. How many species of birds have existed? Bull. Florida State Mus. Biol. Sci. 5(3):39-53.

    Felsenstein, J. 1978. Cases in which parsimony or compatibility methods will be positively misleading. Syst. Zool. 27:401-410.

    Link to this
  32. 32. Heteromeles 3:47 pm 08/28/2012

    “Unfortunately, one can justly argue that the reason that animals with air sacs were around to evolve to gigantism was because their ancestors were the most efficient and biggest things around during the formation of the Siberian traps.”

    You mean – – round about the Permo-Triassic boundary? As in, long before archosauromorphs had evolved pneumaticity?

    Yes, that’s what I meant. I guess my feeble assumption was that you can have a bird/alligator style unidirectional flow of air through lungs without it showing up as air sacs in the bones. Am I massively misinformed about this?

    I should point out that, by this logic, theropods had nothing at all to do with suppressing mammals through the mesozoic. Instead, mammals were kept small by sauropods and especially by ornithiscian dinosaurs outcompeting them as herbivores. Hmmmmmm.

    Link to this
  33. 33. Tayo Bethel 4:31 pm 08/28/2012

    There were probably a number of reasons why mammals werent doing as well as the dinosaurs and other archosaurs in the Mesozoic. Anyone can correct me if I’m wrong, but there dont seem to have been many mammalian herbivores around at any time in the Mesozoic, so being out-competed by dinosaurs as herbivores isnt too unlikely.

    Link to this
  34. 34. BilBy 6:45 pm 08/28/2012

    Darren, re: creationists on this site. One of the nice things about this site is the lack of them; having said that, there were some brilliant, incisive and very informative and clear responses to Bill Crofut’s comments on bird evolution last time he showed up. It’s your site of course, but I would suggest only banning them if they come here to say something like “god rools, yer all going to hell! Dum evilutionists” etc etc. BUT – your page, your rules. I’ll keep coming back whatever :)

    Link to this
  35. 35. Dartian 3:09 am 08/29/2012

    Tayo:
    there dont seem to have been many mammalian herbivores around at any time in the Mesozoic

    Not true; the multituberculates were present from the Jurassic and throughout the entire Cretaceous (and well into the Cenozoic, too). They weren’t necessarily entirely herbivorous, but their rodent-like gnawing teeth strongly suggest that plant matter was a very significant part of their diets.

    (Whether the multituberculates were ‘mammals’ by definition – that is, whether they were members of crown clade Mammalia – is, of course, another question. The jury is still out on that one.)

    BilBy:
    there were some brilliant, incisive and very informative and clear responses to Bill Crofut’s comments on bird evolution last time he showed up

    I see where you’re coming from, but, as I said before, in my opinion that’s not a sufficient reason to tolerate creationist nonsense here. Every substantial response to a creationist takes some time and effort – and that time and effort could and should be spent on something else instead. Such as discussing actual scientific matters with your peers, for example. If you want to have a debate with creationists, there are plenty of sites on the Internet where you can do that until you’re blue in the face. The single most awfully nice thing about Tet Zoo is that it’s such a reliable on-line sanctuary of sanity. Let’s keep it that way, IMO.

    Link to this
  36. 36. Dartian 3:17 am 08/29/2012

    I might add that – again just IMO – a little more leeway could be allowed BANDits and other crackpots here (emphasis on “a little”). If nothing else, they usually are at least marginally more imaginative than genuine creationists, who are simply boring.

    Link to this
  37. 37. Tayo Bethel 4:49 am 08/29/2012

    Dr Naish:

    Thanks for the reminder. Did you do any posts on multituberculates?

    As for the creationists, I personally feel that they should all be consigned to the dark closets where people keep junk they dont really need or that is just plain embarrassing.

    Link to this
  38. 38. David Marjanović 9:00 am 08/29/2012

    Still, if you start with small theropods and small mammals (as in the Danian Paleocene), mammals can evolve to outcompete theropods if their body size isn’t limited by low [O2]. I think the fossil record pretty clearly demonstrates this, if nothing else.

    I’ve never heard of any jump in [O2] at the K/Pg boundary. And even if there was one, you aren’t talking about “small theropods”, you’re talking about “small toothless theropods with reduced tails and practically functionless fingers”. And then the phorusrhacids bite you anyway.

    parsimony assumes that all branches are short (in terms of expected number of changes)

    True.

    and of equal expected length

    False; that would contradict the following.

    (or, alternatively, that the probability of change in one character tells you nothing about the probability of change in any other characters along the same branch).

    True.

    Of course, large enough violations of these assumptions will produce incorrect trees

    True. That’s why maximum likelihood and Bayesian inference were developed.

    Link to this
  39. 39. John Harshman 9:52 am 08/29/2012

    and of equal expected length

    False; that would contradict the following.

    (or, alternatively, that the probability of change in one character tells you nothing about the probability of change in any other characters along the same branch).

    If you really think about it, these are two ways of saying the same thing. If the branch lengths are all the same, then branch length tells us nothing about whether a character is more likely to change on one branch or another, and the probability of change for one character has no relationship to the probability for another. It’s another way to say that branch length parameters are not meaningful inputs to parsimony.

    That annoying add is still blocking the top of the main page, and there’s no way to make it go away. Since it transfers to the bottom of the page when I “keep reading”, I wonder if this feature could be unintentional. And in fact it happens in Safari but not Firefox. Still, a fair number of people must use Safari, so it’s worth a thought.

    Link to this
  40. 40. Heteromeles 10:29 am 08/29/2012

    I’ve never heard of any jump in [O2] at the K/Pg boundary. And even if there was one, you aren’t talking about “small theropods”, you’re talking about “small toothless theropods with reduced tails and practically functionless fingers”. And then the phorusrhacids bite you anyway.

    That’s kind of the point, as a comparison between the P-T and the K-T boundaries. The weird part about the K-T boundary is that the birds and crocs (sensu latu) didn’t stay dominant through the tertiary. They’d pulled that trick before, if I understand the fossil record, and they certainly were the dominant predators in most (if not all) of the Paleocene.

    Link to this
  41. 41. jayfitzsimmons 10:43 am 08/29/2012

    A bit of an aside from the main article, but could someone clarify what a BANDit is? I’ve seen the term a couple times on paleo blogs and googling it gets me nowhere. e.g., comment above “a little more leeway could be allowed BANDits and other crackpots here”. I assume BANDits are a group of people who subscribe to a theory most people regard as unlikely?

    Link to this
  42. 42. llewelly 12:16 pm 08/29/2012

    BAND == Birds Are Not Dinosaurs.

    It helps to know where to look; if you add the qualifier ” site: http://scienceblogs.com/tetrapodzoology ” to your search Darren’s article is the first hit.

    Link to this
  43. 43. David Černý 2:33 pm 08/29/2012

    @John Harshman:

    I have that. But the estimate of total bird species there is just taken from Brodkorb (1959 — where does the 1960 bit come from?), just like the others.

    Thanks for the info.

    Brodkorb was also writing a long time ago, and many of his assumptions have been superseded, for example his notion that most Cretaceous birds were aquatic. We also get into arguments about species concepts, as there are a good thousand or more extant bird species than when he wrote, apparently.

    Could any of these recent developments improve the reliability of Brodkorb’s calculation, or is his approach fundamentally flawed? Is there any chance that the shift to the phylogenetic species concept could make it easier to estimate the number of species living at any given time?

    @David Marjanović

    False; that would contradict the following.

    It’s true that assumptions of parsimony are hard to track down, because the method is a nonparametric equivalent of several different models (F73, G90, TS97 aka “No Common Mechanism”) with different numbers of parameters. However, Goldman’s (1990) version of the “parsimony model” does have a single branch length parameter (= it assumes that all branch lengths are equal). Because of that, Goldman’s model could be (and was — see Goloboff 2003) viewed as extremely simple, but it’s actually still horribly overparameterized owing to its need to estimate ancestral character states at all internal nodes of a given topology.

    True. That’s why maximum likelihood and Bayesian inference were developed.

    No. Parsimony was developed as an approximation to maximum likelihood: Edwards, one of the co-authors of the paper that introduced the method (Edwards & Cavalli-Sforza 1963), later described in detail why they had turned to minimizing the “amount of evolution” in their effort to find an estimate close to the maximum likelihood solution (Edwards 1996). In fact, Edwards & Cavalli-Sforza (1964) put forward maximum-likelihood estimation of phylogenetic trees long before Felsenstein described long-branch attraction. Bayesian methods were developed for other reasons entirely (statistical efficiency, problems with the bootstrap).

    Refs:

    Edwards AWF 1996 The origin and early development of the method of minimum evolution for the reconstruction of phylogenetic trees. Syst Biol 45(1): 79–91

    Edwards AWF, Cavalli-Sforza LL 1963 The reconstruction of evolution. Heredity 18: 533; Ann Hum Genet 27: 104–5

    Edwards AWF, Cavalli-Sforza LL 1964 Reconstruction of evolutionary trees. 67–76 in Heywood VH, McNeill J, eds. Phenetic and Phylogenetic Classification. London: Systematics Association Pub No. 6

    Goldman N 1990 Maximum likelihood inference of phylogenetic trees, with special reference to a Poisson process model of DNA substitution and to parsimony analyses. Syst Zool 39(4): 345–61

    Goloboff PA 2003 Parsimony, likelihood, and simplicity. Cladistics 19: 91–103

    Link to this
  44. 44. jayfitzsimmons 3:42 pm 08/29/2012

    Thank you Llewelly.

    Link to this
  45. 45. Dartian 5:18 am 08/30/2012

    Tayo:
    Thanks for the reminder.

    Psst. Check again who you’re thanking. Just sayin’… (Although Darren most certainly isn’t the worst possible person one could get confused with.)

    Link to this
  46. 46. David Marjanović 9:31 am 08/30/2012

    If the branch lengths are all the same, then branch length tells us nothing

    …Oh. Yeah. Sorry.

    The weird part about the K-T boundary is that the birds and crocs (sensu lat[o]) didn’t stay dominant through the tertiary. They’d pulled that trick before, if I understand the fossil record, and they certainly were the dominant predators in most (if not all) of the Paleocene.

    They suffered from the K/Pg extinction just like everything else and had to recover just like everything else. I don’t know any birds or even crocodiles of Paleocene age that were really big terrestrial predators; they only start in the Eocene.

    I think the Paleocene was largely devoid of large terrestrial predators. Except for Titanoboa, assuming it was terrestrial.

    Parsimony was developed as an approximation to maximum likelihood

    ~:-| Parsimony was developed up to 1950 by Willi Hennig. Maximum likelihood was beyond his dreams; and even WPGMA hadn’t been invented yet, as far as I know. Edwards? Cavalli-Sforza (a name I recognize, of course)? What???

    Link to this
  47. 47. Heteromeles 10:29 am 08/30/2012

    @David: My understanding of the Paleocene (from http://www.paleocene-mammals.de/ among others) was that the large terrestrial predators were birds (e.g. Gastornis) and crocodiles (e.g. Pristichampsus). My assumption is that Gigantophis and allies were aquatic, possibly taking something of a seal-like niche.

    Link to this
  48. 48. John Harshman 4:48 pm 08/30/2012

    ~:-| Parsimony was developed up to 1950 by Willi Hennig.

    Perhaps true for the wimpiest possible sense of “developed”. What Hennig had was an auxiliary principle that was to be used in the unlikely event that not all characters were congruent and you couldn’t resolve that incongruence logically. It was hardly a formal method, let alone an algorithm. If I recall, there are arguments about whether Hennig’s auxiliary principle was much like parsimony at all. Edwards & Cavalli Sforza came up with the first parsimony algorithm.

    Link to this
  49. 49. David Černý 6:20 pm 08/30/2012

    ~:-| Parsimony was developed up to 1950 by Willi Hennig. Maximum likelihood was beyond his dreams; and even WPGMA hadn’t been invented yet, as far as I know. Edwards? Cavalli-Sforza (a name I recognize, of course)? What???

    It’s another point of contention between cladists (who trace the origins of parsimony back to Hennig) and statistical phylogeneticists (who tend to view Edwards and Cavalli-Sforza as the originators of the method). Farris and others have attempted to derive Wagner parsimony directly from Hennig’s auxiliary principle, but many others don’t find their arguments very convincing. Hennig clarified the connection between character evidence and nodes on a phylogenetic tree, but his works don’t seem to contain any specific method for inferring phylogenies. His only method for resolving character conflicts (when different derived character states support conflicting monophyletic groups) apparently was to go back to the problematic characters and restudy their distribution or polarity. Once they are restudied enough, presumably, no internal conflict should occur in the data anymore. That’s what Felsenstein (1982) called “Hennig’s dilemma” — if the incompatibilities can’t be resolved no matter how hard one tries (“no amount of resequencing may be able to change an offending alanine into a glycine” — Felsenstein 1982:381), the logic of Hennigian argumentation doesn’t provide any criterion for choosing between competing hypotheses. Felsenstein showed that two methods can be used to resolve Hennig’s conflict algorithmically — parsimony and compatibility. It isn’t clear which one would be preferred by Hennig, hence he can’t be considered the author of the former method. Strangely enough, Farris et al. (1970:176) agreed with this before they changed their minds (Farris & Kluge 1986, 1997):

    “Unfortunately [Hennig's statement] is not sufficiently detailed to allow us to select a unique criterion for choosing a most preferable tree.”

    Refs:

    Farris JS, Kluge AG 1986 Synapomorphy, parsimony, and explanatory power. Taxon 35: 298–306

    Farris JS, Kluge AG 1997 Parsimony and history. Syst Biol 46(1): 215–21

    Farris JS, Kluge AG, Eckardt MJ 1970 A numerical approach to phylogenetic systematics. Syst Zool 19: 172–9

    Felsenstein J 1982 Numerical methods for inferring evolutionary trees. Q Rev Biol 57(4): 379–404

    Link to this
  50. 50. David Marjanović 6:27 pm 08/30/2012

    My understanding of the Paleocene (from http://www.paleocene-mammals.de/ among others) was that the large terrestrial predators were birds (e.g. Gastornis) and crocodiles (e.g. Pristichampsus).

    Toward the end, yes, I suppose.

    My assumption is that Gigantophis and allies were aquatic, possibly taking something of a seal-like niche.

    I agree; that’s why I mentioned Titanoboa instead. It’s an actual boid, not a madtsoiid.

    What Hennig had was an auxiliary principle that was to be used in the unlikely event that not all characters were congruent and you couldn’t resolve that incongruence logically.

    Even if all characters are congruent, you need a principle to be able to construct a tree from them… but that’s an even wimpier sense.

    If I recall, there are arguments about whether Hennig’s auxiliary principle was much like parsimony at all.

    Then who were the people who calculated most parsimonious trees by hand for lack of alternatives? I’ve been told that Hennig did that. (Though I’ve also been told he thought homoplasy didn’t actually exist, insisting that if you found any, you had to check the coding of your entire matrix.)

    Anyway, given that Hennig’s book was only translated into English in 1966, I wonder if Edwards & Cavalli-Sforza came up with parsimony independently – and the morphologists didn’t notice because they didn’t follow the molecularist literature (and mostly still don’t).

    Link to this
  51. 51. David Marjanović 6:41 pm 08/30/2012

    Hennig clarified the connection between character evidence and nodes on a phylogenetic tree, but his works don’t seem to contain any specific method for inferring phylogenies.

    For the standards of the time, that was a specific method! Everyone else made phylogenetic trees based on assumptions about how evolution works (Cope’s “rule”, Marsh’s “law”, and so on, especially the “rule” that “specialized” organisms can descend from “generalized” ones but not vice versa) and/or from what could be called parsimony with 2 or 3 characters that were chosen for being “important” or “reliable” according to some hunch. Hennig did away with this nonsense and made phylogenetic trees repeatable – turned phylogenetics from an art into a science.

    Before this became widely adopted, there was the period of phylopessimism when people increasingly realized there is no such thing as a reliable character, that everything is subject to homoplasy; increasingly, they gave up on phylogenetics altogether. This period gave us the trees with bubbles and stippled lines that end in question marks.

    Felsenstein showed that two methods can be used to resolve Hennig’s conflict algorithmically — parsimony and compatibility.

    What would compatibility be?

    “Unfortunately [Hennig's statement] is not sufficiently detailed to allow us to select a unique criterion for choosing a most preferable tree.”

    I have to mention that Hennig wrote like a German philosopher. He cared about writing exactly what he meant, inventing a long list of terms in the process, but evidently not about making himself understandable. The English translation was actually translated back into German because the original is (reportedly) so horrible.

    Link to this
  52. 52. David Marjanović 6:43 pm 08/30/2012

    and/or from what could be called parsimony with 2 or 3 characters

    The combination would be an evolutionary scenario that was used to construct a tree. That was done a lot. Today, trees are instead used to test scenarios.

    Link to this
  53. 53. John Harshman 7:33 pm 08/30/2012

    What would compatibility be?

    Ach, kids these days. Compatibility was for a time the main rival to parsimony. In compatibility analysis, you pick the tree that is perfectly compatible (no homoplasy) with the largest set of characters, and ignore all other characters. Sets of mutually compatible characters are called “cliques”, though I understand the term was originally intended pejoratively. I bet Felsenstein’s book covers all this.

    Hennig’s method merely supposed that each informative character defined a node, so all you had to do was assemble a tree from all the defined nodes. Homoplasy ws assumed not to exist, i.e. to reflect a coding error on your part that ought to have been disposed of before analysis.

    Link to this
  54. 54. David Marjanović 5:38 am 08/31/2012

    In compatibility analysis, you pick the tree that is perfectly compatible (no homoplasy) with the largest set of characters, and ignore all other characters.

    Huh. Unless you have a very small matrix or a very cherry-picked one or lots and lots of correlated characters, the largest such set is going to be tiny – and hardly larger than the several next largest sets.

    I don’t know how Hennig chose characters for inclusion in his data matrices – did he have any criteria other than that they had to be parsimony-informative? If yes, that suggests compatibility, because it suggests trying to identify that largest compatible set before analysis and ignoring all other characters.

    Link to this
  55. 55. David Černý 7:10 am 08/31/2012

    Then who were the people who calculated most parsimonious trees by hand for lack of alternatives? I’ve been told that Hennig did that.

    I don’t think so, and even some progressive cladists (Ed Wiley, for example) would agree with me on this:

    “As a summary, there are four methods of phylogenetic analyses that do not rely on phenetic (distance estimates) methodology: (1) classical Hennigian argumentation, (2) parsimony analysis, (3) likelihood analysis, and (4) Bayesian inference. Of the four, Hennig is known to us only to have used Hennigian argumentation.”

    Wiley et al. 2011:11

    BTW, speaking of lack of alternatives… by 1966, Edwards and Cavalli-Sforza had a working Fortran-based program for calculating “minimum-length” trees (EVOMIN).

    For the standards of the time, that was a specific method!

    You’re right. What I meant was that he didn’t have any optimality criterion, unlike Edwards and Cavalli-Sforza (who actually invented three different optimality criteria in two years). However, he had a method, and it worked reasonably well provided that there was no internal conflict in the data, even though it was inefficient.

    Hennig did away with this nonsense and made phylogenetic trees repeatable – turned phylogenetics from an art into a science.

    That’s one possible viewpoint. The other one, advocated by Felsenstein and others, is that phylogenetics was turned into a science by Sokal and Michener in the late 1950s through UPGMA. I’m inclined to agree — UPGMA is an algorithmic method capable of dealing with conflicts among characters and trivial enough to be computerized, which are qualities that Hennigian argumentation seems to lack.

    Before this became widely adopted, there was the period of phylopessimism when people increasingly realized there is no such thing as a reliable character, that everything is subject to homoplasy; increasingly, they gave up on phylogenetics altogether.

    Of course, the Modern synthesis and the emphasis on population genetics and speciation mechanisms didn’t exactly help either…

    If yes, that suggests compatibility, because it suggests trying to identify that largest compatible set before analysis and ignoring all other characters.

    Then again, he might have believed that there are no incompatibilities among characters if you study them really really hard. In morphology, such an idea makes a little more sense than in molecular biology.

    Ref:

    Wiley EO, Chakrabarty P, Craig MT, Davis MP, Holcroft NI, Mayden RL, Smith WL 2011 Will the real phylogeneticists please stand up? Zootaxa (2946): 7–16

    Link to this
  56. 56. Jerzy v. 3.0. 8:07 am 08/31/2012

    Re: bird species
    One caveat is the theory that there were recently 1000s of extinct bird species narrowly endemic to small islands in Pacific. So the total number of bird species may be much influenced by islands.

    Re: sauropods and air sacs.
    I understand is that birdlike respiration was lacking or degenerated in ornithischians, but they nevertheless took place of sauropods.

    I think sauropod dominance and fall may have more to do with other factors. Possibly constant evolution of predators towards more speed and activity. It made juvenile sauropods too easy a meal, and favored groups which protect their young: ornithischians, birds and mammals.

    Link to this
  57. 57. David Marjanović 10:05 am 08/31/2012

    That’s one possible viewpoint. The other one, advocated by Felsenstein and others, is that phylogenetics was turned into a science by Sokal and Michener in the late 1950s through UPGMA. I’m inclined to agree — UPGMA is an algorithmic method capable of dealing with conflicts among characters and trivial enough to be computerized, which are qualities that Hennigian argumentation seems to lack.

    Today’s computer programs for phylogenetic analysis are of course based on those for phenetic analysis; I won’t try to discuss this contribution away.

    It’s serendipitous, however. Isn’t Sokal the one (or is it Sneath?) who plainly couldn’t imagine why anyone would ever want to reconstruct a phylogeny? “If you had a phylogeny, what would you do with it?” The idea behind UPGMA was to have an objective, easy-to-use way to translate a distance matrix into a hierarchy for the purpose of generating a stable classification*. It was not to reconstruct phylogeny.

    * Of course, at this point I have to mention that Hennig similarly viewed phylogenetics only as the means for the end of making a classification, something that has been largely abandoned.

    Then again, he might have believed that there are no incompatibilities among characters if you study them really really hard.

    Apparently he did.

    In morphology, such an idea makes a little more sense than in molecular biology.

    Indeed. The matrix I’m working on gives trees with CIs well below 0.3.

    One caveat is the theory that there were recently 1000s of extinct bird species narrowly endemic to small islands in Pacific. So the total number of bird species may be much influenced by islands.

    One species of flightless rail for every small island, and several for every big one.

    I understand is that birdlike respiration was lacking or degenerated in ornithischians

    All we know is that it didn’t leave traces on the bones in ornithischians.

    I think sauropod dominance and fall may have more to do with other factors.

    Fall? What fall? That only happened in North America, during a sea level highstand.

    Possibly constant evolution of predators towards more speed and activity.

    Evidence?

    Link to this
  58. 58. David Marjanović 10:05 am 08/31/2012

    I forgot to thank you for the reference to Wiley et al. (2011); I’ll read it.

    Link to this
  59. 59. David Černý 2:52 pm 08/31/2012

    It’s serendipitous, however. Isn’t Sokal the one (or is it Sneath?) who plainly couldn’t imagine why anyone would ever want to reconstruct a phylogeny? “If you had a phylogeny, what would you do with it?” The idea behind UPGMA was to have an objective, easy-to-use way to translate a distance matrix into a hierarchy for the purpose of generating a stable classification*. It was not to reconstruct phylogeny.

    You are right again, partially. Thus Spoke Felsenstein (2004) about the tree in the seminal 1957 paper by Michener and Sokal:

    “[...] interpretation as a phylogeny was made by Michener; Sokal saw it as a classification that did not necessarily have any validity as a phylogeny.”

    Apparently, there was no single idea behind UPGMA and the two authors of the algorithm had rather different opinions about what it was supposed to do.

    One caveat is the theory that there were recently 1000s of extinct bird species narrowly endemic to small islands in Pacific. So the total number of bird species may be much influenced by islands.

    One species of flightless rail for every small island, and several for every big one.

    Steadman (1995) actually tried to calculate the number, his conclusion being that there were about 2,000 flightless rail species prior to the colonization of Oceania. It follows that the total number of recently extinct island bird species must have been even higher.

    All we know is that it didn’t leave traces on the bones in ornithischians.

    … with the possible exception of Delapparentia. I’d love to see more research on this taxon.

    I forgot to thank you for the reference to Wiley et al. (2011); I’ll read it.

    You’re welcome. :) It’s an interesting paper — Wiley et al. wrote it as a response to a group of fish systematists* who insist that optimization is bad, any algorithmic phylogenetics is really “phenetics”, and the fossil of pattern cladism must be resurrected as the only right way to infer phylogenies. In their view, parsimony is phenetics, too, because its results can change with the addition of new taxa (yes, really), which in their opinion implies “a misunderstanding of character homology”. Wiley et al. responded by repeating the whole history of phylogenetic methods for them, showing why people progressed from Hennigian argumentation to parsimony and from parsimony to probabilistic methods, and demonstrating some interesting facts in the process (they re-ran Gauthier’s 1988 amniote analysis with Bayesian inference and ancestral state reconstruction methods in order to show that probabilistic methods tend to group by synapomorphies).

    *They are actually systematists working on fishes, not systematists who are fishes. It’s a subtle but important distinction.

    Refs:

    Felsenstein J 2004 Inferring Phylogenies. Sunderland MA:
    Sinauer

    Steadman DW 1995 Prehistoric extinctions of pacific island birds: biodiversity meets zooarchaeology. Science 267(5201): 1123–31

    Link to this
  60. 60. David Marjanović 10:07 am 09/1/2012

    Apparently, there was no single idea behind UPGMA and the two authors of the algorithm had rather different opinions about what it was supposed to do.

    I see.

    *They are actually systematists working on fishes, not systematists who are fishes. It’s a subtle but important distinction.

    Oh, in the other case they’d be systematist fish.

    And now is the moment for waffling about fish paraphyly! Or were they the authors of the awful “a fish is a fish” paper? Well, I’ll find out. Now just allow me a few hours to come to term with the concept of zombie pattern cladists. *epic facepalm*

    they re-ran Gauthier’s 1988 amniote analysis with Bayesian inference and ancestral state reconstruction methods in order to show that probabilistic methods tend to group by synapomorphies

    Awesome!

    Link to this

Add a Comment
You must sign in or register as a ScientificAmerican.com member to submit a comment.

More from Scientific American

Scientific American Holiday Sale

Give a Gift &
Get a Gift - Free!

Give a 1 year subscription as low as $14.99

Subscribe Now! >

X

Email this Article

X