About the SA Blog Network

Tetrapod Zoology

Tetrapod Zoology

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

The other peacock

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

Email   PrintPrint

Awesome male Green peacock, though lacking train (captive individual at Tierpark Berlin), photo by Markus Bühler, used with permission.

Mention ‘peacock’ (or ‘peafowl’) and the vast majority of people will think of Pavo cristatus, the mostly Indian, blue-plumaged Indian peacock. Distinctive features of the males of this species include a face marked with black and white stripes, a crest composed of wire-like shafts tipped with blue, fan-shaped tufts, orange-brown primary feathers, dark-brown-and-white stripes on the scapulars and wing coverts, and of course an iridescent blue neck and breast and a massive ‘train’ formed of uppertail coverts marked with eyespots. Females have white bellies, brown backs and greenish necks, and of course lack the long ‘trains’ of males.

Photo by Markus Bühler, used with permission.

But there’s another peafowl: the Green peacock P. muticus, also sometimes called the Burmese peacock. Historically occurring across the woodlands, forests and montane grasslands of southern Asia from north-eastern India eastwards to Vietnam and south to Malaysia and Java (but with no recorded presence on Sumatra or Borneo), it has undergone a massive and worrying decline since the middle of the 20th century and is now rare and restricted to a shrinking number of isolated populations. A reliance on regular sources of water and use of forest edges and cultivated areas may bring it regularly within range of humans and its decline is predominantly due to hunting and habitat loss.

Something we have to sort out up front here is the apparent interchangeability of the terms ‘peacock’ and ‘peafowl’. No, the latter term is not used specifically for females, nor is the former exclusively used for males. It seems that both can be used in the singular, but that only the latter can be used in the plural. My just-arrived-at convention is that we speak of ‘peacocks’ when addressing members of a single taxon (e.g., this article is about the Green peacock) but use ‘peafowl’ when referring to peacocks of more than one taxon (e.g., there are two living species of peafowl).

Green peacocks typically have a more erect-necked look than Indian peacocks and are larger and taller overall. They’re also far less sexually dimorphic: the female is about the same in appearance as the male but for her absence of the train. That reduced amount of sexual dimorphism may tell us something about sociosexual behaviour – read on. As you can see from the photos used here, the Green peacock really is a spectacular, striking and large bird. Indeed, with a total length of up to 3 m for a train-bearing male, this is the world’s longest gamebird. Large individuals can weight 5 kg.

Displaying male Green peacock. Note the naked grey, pale blue and yellow facial skin. Photo by Arddu but from wikipedia where licensed under Creative Commons Attribution 2.0 Generic license.

The many kinds of Green peacock

Photo by Markus Bühler, used with permission.

Not all populations are alike within this species; three subspecies have typically been considered valid (Madge & McGowan 2002) but there are suggestions that this classification is too simplistic and that several additional forms need to be recognised. P. m. muticus is endemic to Java and is the greenest and most brightly coloured of the three. P. m. imperator is actually the best known one: it’s the one that used to occur throughout southeast Asia and adjacent Burma and China. The head crest is formed from narrow, tufted feathers and the neck, breast and flanks are mostly green but with black centres; metallic blue or blue-green feathers are present on the head and parts of the neck and breast too. Finally, P. m. spicifer is the dullest of the three, with near-black secondaries and wing coverts, and slightly blue neck and breast feathers. This more pronounced bluish-ness makes or made it the subspecies most similar to the Indian peacock. Reported only from north-east India and north-west Burma, it may perhaps be extinct (Madge & McGowan 2002).

Additional subspecies recognised by some authors include P. m. yunnanensis from Yunnan, China, P. m. annamensis from south-east Asia and Yunnan and P. m. javanensis from Java. If the species really does include this many distinct taxa, it is (or was) evidently diverse genetically.

A long history of peafowl

Genetic data suggests that Indian and Green peafowl diverged about 3 million years ago (Ouyang et al. 2009). If correct, this means that both are old and substantially distinct; it also raises the question as to how old those Green peacock subspecies are, and already there are suggestions from some in the pheasant world that the Green peacock is actually a species complex and that taxonomic revision is appropriate.

Intriguingly, fossil peafowl from Europe have been said to be more similar to Green peacocks than to Indian peacocks (Mourer-Chauviré 1989). However, this is almost certainly more to do with the fact that both the fossil specimens and the Green peacock share large size and hence are similarly proportioned when it comes to those parts of anatomy most frequently preserved as fossils, like tarsometatarsi and spur bones.

The sorts of fossils we're often dealing with when looking at fossil peafowl and other gamebirds: tarsometatarsi. In this figure (from Mourer-Chauviré 1989), the large tarsometatarsus is that of Pavo bravardi; the smaller one is a domestic chicken.

As we saw in a recent article, fossil peafowl are known from the Miocene and Pliocene of Europe; they were also present in Europe in the Early Pleistocene too (Mourer-Chauviré 1989). Pavo bravardi (originally described in 1849 as a junglefowl) from France, Greece, Bulgaria, Moldavia and Ukraine seems to have been a cool-adapted animal, and indeed its fossils are, in Bulgaria, found in association with those of ptarmigan (Boev 1998). P. moldavicus from the Pliocene of Moldavia (Bocheński & Kurochkin 1987) is probably synonymous with P. bravardi (Mourer-Chauviré 1989).

Another European member of PavoP. aesculapi – is known from the Late Miocene of Greece and elsewhere in Europe. There’s also a Kenyan fossil Pavo (as yet unnamed to species level) from the Pliocene.

Lophogallus naranbulakensis Zelenkov & Kurochkin, 2010: this is the holotype - the distal end of the humerus. Not all fossil gamebirds are like this, honest. From Zelenkov & Kurochkin (2010).

We don’t yet know how these fossil peafowl were related to the modern ones, but they do show that peafowl of the modern sort – that is, members of the clade Pavo – have been around since the Late Miocene at least. Even better, some entirely extinct phasianids might also be members of the peafowl lineage, in which case the group has a record stretching back to the Middle Miocene. The best known of these is Miophasianus altus, known from widespread localities across Europe (Sánchez Marco 2006) and often said to be one of the largest of fossil gamebirds. Cheneval (2000) was first to  suggest that it might be a peacock. Linquornis gigantis, also from the Middle Miocene but this time from China, is also large (similar to living Pavo) and sometimes said or implied to be peacock-like (Zelenkov & Kurochkin 2010) while Lophogallus narabulakensis from the Middle Miocene of Mongolia was also large and peacock-like (Zelenkov & Kurochkin 2010).

The great problem with so many Cenozoic fossil birds is that there has – so far – been little effort to investigate their relationships empirically: conclusions are generally done by reference to overall similarity (indeed, quite a few Cenozoic bird specialists are avowed anti-cladists). Furthermore, the fossil taxa themselves are frequently represented by scrappy remains (the fossil peafowl and peafowl-like phasianids mentioned here, for example, are named for isolated or partial limb bones, not whole or semi-whole skeletons), and hence are not ideal subjects for phylogenetic investigation anyway. More fossils are always needed.

However, what material we have indicates that peafowl have a record going back well into the Miocene – an observation consistent with phylogenetic trees that show pavonines (the members of the pheasant group that includes peacock-pheasants, argus pheasants and peafowl) diverging from an ancestor that later gave rise to northern pheasants, grouse, turkeys and other lineages (e.g., Kriegs et al. 2007, Kimball et al. 2011): in other words, diverging early on in the history of the group.

Time-calibrated gamebird phylogeny, from Kan et al. (2010). Note the really ancient divergence dates for many lineages (megapodes diverging in the Cretaceous?) - can this be right?

Incidentally, the molecular clocks and divergence dates suggested for various gamebird lineages have been controversial and sometimes faulty due to the misinterpretation of fossils used as calibration points (see Ksepka 2009). I was surprised to see one recent suggestion positing the divergence time for pavonines and their presumed closest relatives (junglefowl and kin in the phylogeny concerned) to date to the Eocene-Oligocene boundary (Kan et al. 2010). That’s much older than the fossils suggest, and indeed inconsistent with the gamebird fossil record as a whole: crown-galliforms aren’t present until the end of the Oligocene. I think that results such as this are inaccurate and presumably caused by incorrect assumptions about molecular clock rates.

Which social system for you, Green peacock?

Like the better-known Indian peacock, the Green peacock is mostly assumed to be polygamous or polygynous, with males holding harems of several females and advertising their presence with loud calls that are apparently less piercing than those of the Indian peacock.

Another view of that captive specimen. Photo by Markus Bühler, used with permission.

However, the species apparently does not form leks like the Indian peacock and there are suggestions, stemming from observations made in captivity, that these birds might actually be monogamous and with males performing an extensive role in parental care. This would explain the strong similarity present between males and females and wouldn’t be entirely discordant with what we already know about pavonines, since the Congo peacock Afropavo congensis is apparently monogamous.

It used to be thought that the presence of well-developed spurs in gamebirds was tidily correlated with the mating system: the Green peacock has large spurs, suggestive of polygyny or polygamy. However, at least one study has found that the correlation supposed to exist here is actually ambiguous (Sullivan & Hillgarth 1993). Clearly, better data on wild Green peacocks is needed before we can state anything with confidence about the mating and breeding behaviour of this species.

All in all, a remarkable bird: one that we really need to understand better, and one that we really need to see conserved before it becomes too late. Many thanks for Markus Bühler for providing the great photos used here.

For previous Tet Zoo articles on gamebirds, see…

Refs – -

Bocheński, Z. & Kurochkin, E. N. 1987. New data on Pliocene phasianids (Aves, Phasianidae) of Moldavia and s. Ukraine. Acta zoologica Cracoviensia 30, 81-96.

Boev, Z. 1998. Peafowls (g. Pavo Linnaeus, 1758) and ptarmigans (g. Lagopus Brisson, 1760): an unique coexistence in North Bulgaria over 3 m.y. ago. Biogeographica 19, 220-222.

Cheneval, J. 2000. L’avifaune de Sansan. Mémoires du Muséum national d’Histoire naturelle 183, 321-388.

Kan, X.-Z., Li, Z.-F., Lei, Z.-P., Chen, L., Gao, H., Yang, Z.-Y., Yang, J.-K., Guo, Z.-C., Yu,. L,. Zhang, L.-Q. & Quan, C.-J. 2010. Estimation of divergence times for major lineages of galliform birds: evidence from complete mitochondrial genome sequences. African Journal of Biotechnology 9, 3073-3078.

Kimball, R. T, St. Mary, C. M. & Braun, E. L. 2011. A macroevolutionary perspective on multiple sexual traits in the Phasianidae (Galliformes). International Journal of Evolutionary Biology 2011, Article ID 423938, 16 pp., doi: 10.4061/2011/423938

Kriegs, J. O., Matzke, A., Churakov, G., Kutizin, A., Mayr, G., Brosius, J. & Schmitz, J. 2007. Waves of genomic hitchhikers shed light on the evolution of gamebirds (Aves: Galliformes). BMC Evolutionary Biology 2007, 7:190 doi:10.1186/1471-2148-7-190

Ksepka, D. T. 2009. Broken gears in the avian molecular clock: new phylogenetic analyses support stem galliform status for Gallinuloides wyomingensis and rallid affinities for Amitabha urbsinterdictensis. Cladistics 25, 173-197.

Madge, S. & McGowan, P. 2002. Pheasants, Partidges & Grouse, Including Buttonquails, Sandgrouse and Allies. Christopher Helm, London.

Mourer-Chauviré, C. 1989. A peafowl from the Pliocene of Perpignan, France. Palaeontology 32, 439-446.

Ouyang, Y.-n., Yang, Z.-y., Li, D.-l., Huo, J.-l., Qian, K. & Miao, Y.-w. 2009. Genetic divergence between Pavo muticus and Pavo cristatus by cyt b gene.  Journal of Yunnan Agricultural University 24, 220-224.

Sánchez Marco, A. 2006. Miophasianus and Palaeoperdix (Galliformes, Aves) from three Miocene localities of Spain. Estudios Geológicos 62, 249-256.

Sullivan, M. S. & Hillgarth, N. 1993. Mating system correlates of tarsal spurs in the Phasianidae. Journal of Zoology 231, 203-214.

Zelenkov, N. V. & Kurochkin, E. N. 2010. Neogene phasianids (Aves: Phasianidae) of Central Asia: 3. Genera Lophogallus gen. nov. and Syrmaticus. Paleontological Journal 44, 328-336.

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 He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at! Follow on Twitter @TetZoo.

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

Rights & Permissions

Comments 36 Comments

Add Comment
  1. 1. comandantedavid 10:26 am 01/17/2013

    When the farm I worked at bought two peafowl, we used “peacock” for the male, “peahen” for the female, and “peafowl” for the pair, following usage for chickens and guinea fowl. But then I guess that leaves out the neuter singular: “peabird”?

    Link to this
  2. 2. comandantedavid 10:29 am 01/17/2013

    Addendum: apparently “fowl” is singular as well as plural, so there you have it.

    Link to this
  3. 3. John Harshman 11:56 am 01/17/2013

    Molecular clocks can obviously have calibration problems when there’s a lot of variation among taxa in evolutionary rate. Sprinkled through the tree are various taxa with unusually high rates of molecular evolution, for reasons not entirely clear. Phasianidae appears to be one of those taxa. An external calibration point might make divergences in Phasianidae look older than they are; then again, an internal calibration point shouldn’t have that problem. Toss in a Bayesian analysis with multiple internal and external calibrations, and I have no idea what would happen.

    But I see no problem with the divergence between megapodes and other galliforms being Cretaceous. Given Vegavis, how can that be considered unlikely? For what it’s worth, a very simple and stupid (non-parametric rate smoothing) calibration of the Early Bird tree puts the divergence at 65ma.

    Link to this
  4. 4. David Marjanović 12:09 pm 01/17/2013

    Toss in a Bayesian analysis with multiple internal and external calibrations, and I have no idea what would happen.

    The more calibrations, the better; the more evenly distributed the calibrations (internal and external, young and old), the better; if all calibrations (or all but the root) are minimum ages only, that pushes everything towards older dates. Brochu 2004, 2006; me & Laurin 2007.

    But I see no problem with the divergence between megapodes and other galliforms being Cretaceous.

    There’s a nice Hennig comb of stem-galliforms from the Paleogene, while crown-galliforms are absent till nearly its end. On the anseriform side, we have presbyornithids – crown-anseriforms – all over the Eocene and Paleocene and even into the Cretaceous (Teviornis if it’s still accepted; I forgot).

    Link to this
  5. 5. Dartian 12:24 pm 01/17/2013

    Brilliant follow-up to the ocellated turkey article, Darren! :)

    This more pronounced bluish-ness makes or made it the subspecies most similar to the Indian peacock. Reported only from north-east India and north-west Burma

    Is/was it sympatric with cristatus? If so, could the similarity in appearance be due to hybridization?

    Link to this
  6. 6. naishd 12:40 pm 01/17/2013

    John (comment 3): what you say is fair enough, but I suppose I am especially sceptical because a divergence of this sort (positing megapodes and hence other crown-galliforms in the Cretaceous) is really inconsistent with the fossil record – there are quite a few Eocene and early Oligocene galliforms, and all are outside the crown. Crown-galliforms don’t appear until the Late Oligocene. As for anseriforms, Teviornis is an alleged Late Cretaceous stem-form but its identification has been doubted; that leaves Vegavis and there are doubts about that too (patience, patience). A few Paleogene anseriforms (alleged screamers and anseranatids) seem to be part of the crown.

    An in-preparation project leads me to doubt some of the stuff I’ve said before around the timing of galloanserine (and neornithine) origins.


    Link to this
  7. 7. BrianL 12:59 pm 01/17/2013

    I’m glad you raised the point of how reliable molecular trees for birds are, along with the general scrappiness of their fossil record. I’m not a paleontologist myself, so I suppose I’ll have to trust those that are, but I still find it rather discomforting to see so many identifications of fossil birds being based on such scrappy material as isolated tibiotarsa or tarsometatarsals. From my outsider’s perspective, it would seem that such bones are prone to convergence and may well be misleading when it comes to allying them with extant taxa.

    I can only guess at the amount of undeliberate misinformation presented about the avian fossil record when non-cladistic approaches are apparently as common as Darren says they are. As such, I now have an explanation for Oligocene or even older fossils being assigned to modern genera of birds (For example, to *Pandion*,* Haliaaetus* and *Pelecanus*.) Do we really expect a fossil that old to lie inside the most inclusive clade comprising its modern species? To me, this screams fossils being assigned to a modern genus based on plesiomorphic characters, a clear result of not using cladistics.

    What I find even worse is pure biogeographical grounds being used to assign a fossil belonging likely allied to a modern clade to a modern genus. A good example of this I find *Conuropsis fratercula*. A single tarsometatarsus of a Miocene Nebraskan parrot being assigned to modern (if recently extinct) *Conuropsis*. If there was any reason beyond biogeography (both were parrots occurring far within the borders of what we now call the United States) for this assignment, I have never found out. Yet, such fossils are being used to ‘prove’ how old a genus is and calibrate trees, whereas it may well be nothing but a red herring in that regard.

    And last, can molecular studies please stop using continental break-ups as calibration for avian evolution? I can only facepalm when I see, for example, acanthisittid divergence from other passeriforms being dated to about 80 million years, because that’s when Zealandia split off from Australia. Birds as a whole are great dispersers, so vicariance is not likely to be a good explanation for present day distribution patterns. What’s more, we should expect the vast majority (if not the entirety) of living avian clades below the Neoaves-Galloanserae divergence to have evolved within the Cenozoic so continental drift in the Mesozoic should not be considered vital or even important in explaining their modern distribution.

    All in all, I feel as if our understanding of neornithean evolution is muddled, if not hindered, by false assumptions, shoddy calibration of data and often poor fossils being given far more importance than they should deserve with non-cladistic approaches being an important reason for that.

    I don’t get the impression that the same things apply to the study of mammalian fossils and phylogeny. Is that just me or is it truly the case?

    Apologies for ranting about this when the subject should be peacocks.

    Link to this
  8. 8. vdinets 1:03 pm 01/17/2013

    There is something like a lek in Baluran Nat’l Park, Java, where you often see a few males displaying within view of each other in the morning, but that can be due to the presence of a watering hole. Nothing like leks in Cat Tien in Vietnam, as far as I know.

    Link to this
  9. 9. Hydrarchos 1:34 pm 01/17/2013

    I knew that green peafowl existed (my understanding of the terminology was the same as comandantedavid’s), but for some reason had thought they were smaller and less spectacular than blue peafowl. Is there some (perhaps behavioural?) reason why it was cristatus rather than muticus that become the commonly captive-bred/”ornamental” species?

    This specimen really looks much leaner, longer-legged and more “alert” than most captive blue peafowl – with the “scale” pattern on the neck and the long, stiff tail it really makes me think of heraldic dragons and the popular imagination of “scaly” non-avian dinosaurs. The thought that similar birds could have been stalking around Europe relatively recently makes me think of some Welsh (and possibly also Eastern European?) legends I have read about of feathered, winged “serpents” with bright, iridescent colours (I think they were said to have both scales and feathers) that were regarded as a “pest” to livestock (like, say, raptors or foxes). I think I have seen some cryptozoologists speculating that these could have been “inspired by” an extinct species of long-tailed pheasant (similar to Reeves’ pheasant), but a peafowl species could be an even better fit (as they are more predatory than most galliforms, IIRC). Is there any evidence that they might have survived long enough to coexist with humans?

    Link to this
  10. 10. vdinets 1:54 pm 01/17/2013

    Hydrarchos: blue peacock is a common bird of agricultural landscapes in many parts of India. Why people in India have been more successful in preserving and domesticating wild fauna in densely populated areas than people in SE Asia is a really interesting question. Note that elephants and chickens were apparently first domesticated in India, too.

    Link to this
  11. 11. naishd 5:26 pm 01/17/2013

    Such great comments, I don’t have time to respond properly. Hydrarchos (comment 9): you are right, I think, about the ‘dragony’ nature of the Green peacock. Indeed, one name used for some forms – apparently based on a local name – is Dragonbird.

    As for the cryptozoological suggestion that a large, long-’tailed’ pheasant of some sort might explain European sightings of dragons, ‘winged snakes’ and such, it would require the persistence of a giant, peacock or peacock-like form into modern times, and this is hard to take seriously given what we presently know about the zooarchaeological record. I think it’s also one of those cases, common in cryptozoology, where people count the hits but ignore the misses – so, the creatures concerned had long tails, were brightly coloured and could fly, but were any other peacock-like, pheasant-like or gamebird-like features ever mentioned? (like beaks, spurs, feathers) I don’t think so.


    Link to this
  12. 12. naishd 5:30 pm 01/17/2013

    vdinets (comment 10): regarding the domestication of chickens… it’s complicated, but south-east Asia – Thailand especially – is probably the centre of their domestication, not India. At least this is what the phylogeography suggests.


    Link to this
  13. 13. vdinets 7:30 pm 01/17/2013

    Darren: “This study was later found to be based on incomplete data, and recent studies point to multiple maternal origins, with the clade found in the Americas, Europe, Middle East, and Africa, originating from the Indian subcontinent, where a large number of unique haplotypes occur.” See: Liu et al. (2006) Multiple maternal origins of chickens: Out of the Asian jungles. Molecular Phylogenetics and Evolution 38 (1): 12–19.

    Also, note that domestic chicken have some genes from grey junglefowl, which is endemic to India.

    Eriksson J. et al. (2008) Identification of the Yellow Skin Gene Reveals a Hybrid Origin of the Domestic Chicken. PLoS Genet January 23, 2008

    Link to this
  14. 14. Hydrarchos 8:59 pm 01/17/2013

    I did a bit of searching for the flying serpent thing, and found what must have been the source of my memory: – although that post does seem to be full of even more unfounded speculation than the original idea…

    (also it seems that hybrids between Green and Blue peafowl are common in captivity, and fertile enough to cause difficulties identifying “pure” breeding lines of the Green species, with “pure” individuals of subsp. spicifer being the hardest to distinguish from hybrids, perhaps giving some credence to spicifer having some cristatus ancestry, according to a PDF that I now can’t re-find the link to…)

    Link to this
  15. 15. FrankLegend 4:02 am 01/18/2013

    In fact Green Peafowl may be split into 5 to 7 distinct species with their own subspecies. The data is preliminary and has not been published yet. However, Chinese scientists also supported the theory there are two forms in Yunnan.

    @Hydrarchos. In fact it’s the total opposite. Spicifer along with one of the forms in Yunnan (which Darren terms as yunnanensis, but some authors refer to it as antiqus) are the least derived Green Peafowl. The two are much larger than other peafowl, the Yunnan bird being the larger of the two. It may represent a living population of Pavo bravardi.

    I believe Imperator is the closest related to cristatus. There is a critically endangered variation of Imperator found in Tonkin and Hue Vietnam as well as Laos that looks even more like a hybrid. West Thailand is an area where the three forms, annamensis, imperator and spicifer have interbred. Imperator in itself is a descendant of annamensis and spicifer integrading. A form intermediate between annamensis and imperator is found in Northern Thailand. Also the Malaysian and the Javanese birds were not the same.

    However, more clearly needs to be done to investigate these birds.

    Link to this
  16. 16. naishd 4:20 am 01/18/2013

    vdinets (comment 13): like I said, it’s complicated – my point was that India should probably not be thought of as the sole centre of domestication as it used to; rather that south-eastern Asia and other areas (India included) contributed to the story. Liu et al. (2006) say southwest and southern Asia; more recently, Kanginakudru et al. (2008) concluded that “the domestication of chicken has occurred independently in different locations of Asia including India”.


    Ref – -

    Kanginakudru, S., Metta, M., Jakati, R. D. & Nagaraju, J. 2008. Genetic evidence from Indian red jungle fowl corroborates multiple domestication of modern day chicken. BMC Evolutionary Biology 8, 174 doi:10.1186/1471-2148-8-174

    Link to this
  17. 17. Jerzy v. 3.0. 6:44 am 01/18/2013

    @BrianL – I would agree with most of what you say.

    @9 Green Peacock is sensitive to cold and internal parasites. In fact, many aviculturalists keep so-called “emerald peacocks” or hybrids looking like muticus but have much of resistance of cristatus. I can only wonder if captive line of “pure” Green Peacock is really pure.

    That Blue Peacock is cold-hardy is not surprising, if you remember that semi-desert of NW India can be really chilly in winter.

    Link to this
  18. 18. David Marjanović 9:18 am 01/18/2013

    that leaves Vegavis and there are doubts about that too (patience, patience).


    I can only guess at the amount of undeliberate misinformation presented about the avian fossil record when non-cladistic approaches are apparently as common as Darren says they are. As such, I now have an explanation for Oligocene or even older fossils being assigned to modern genera of birds (For example, to *Pandion*,* Haliaaetus* and *Pelecanus*.) Do we really expect a fossil that old to lie inside the most inclusive clade comprising its modern species?

    …Wait. Nobody said all genera have to be their own crown-groups. At least some of these assignments were clearly made by people who conceptualized the genera in question as total groups, and most had (if anything) something more like an apomorphy-based definition in mind.

    Also, Pandion has a single extant species, so any other species assigned to that genus necessarily lies outside the crown-group.

    Link to this
  19. 19. John Scanlon FCD 9:35 am 01/18/2013

    Why am I thinking of brachiosaurs now?

    Link to this
  20. 20. Heteromeles 10:11 am 01/18/2013

    Got to point out, with a group that has such porous breeding barriers, using genes to figure out where a species was first domesticated is going to be problematic.

    Why? Because wild animals will breed with domesticated animals.

    In some cases, farmers will deliberately breed their animals (and plants) with wild relatives to improve their hardiness. This has been documented for everything from corn and sunflowers to pigs and dogs, and I’d be really surprised if it didn’t happen for chickens, turkeys, peacocks, and other fowl.

    When we look at domestic animals traits, what we may be seeing are morphological features that correlate with a) tameness, b) disease resistance, and c) productivity, since these are things that farmers select for. The other genes don’t matter as much, and they could be a hodge-podge of relics from all ancestors.

    Couple this with the innate propensity to make stories out of random patterns, and I’d suggest everyone needs to be careful when writing the story of where and how clades became domesticated.

    Link to this
  21. 21. BrianL 10:44 am 01/18/2013

    @David Marjanovic:
    You are right in saying this, of course. I should have said more about apomorphy-based classification in modern genera, but feared that this would make my comment even longer than it already was.

    It is this apomorphy-based classification that, to my mind, leads to plesiomorphic characters being used to place such taxa in modern genera thanks to the general paucity of neornithean fossil record. With *Pelecanus*, for example, Oligocene pelicans show all evidence of having the same extreme beaks (presumably complete with gular pouches) as modern species do and have been placed in the modern genus *Pelecanus* for that reason. Yet, with only a single modern (though presumably quite old) genus around and a very meagre fossil record, we really can’t say wether the pelican’s beak was present in other, extinct pelicans or in their concestor with *Balaeniceps* or even further down the tree (nor would I be surprised if the Eocene balaenicipitid *Goliathia* was actually basal to (*Balaeniceps* + *Pelecanus*). The lack of possibility for meaningful comparison to other fossil taxa means that we should not a priori assume we’re dealing with a synapomorphy. This makes such classifications very risky in my view, as you might well be using a plesiomorphic character as diagnostic for your classification. With non-cladistic approaches being so common within paleornithology,this has severe implications for the field as a whole, I fear.

    Undoubtedly, you’re right in pointing out that apomorphy-based classifications should not be judged by cladistic principles, but I do feel that modern genera should be considered crown clades if we wish to meaningfully discuss their evolutionary history and how far they go back in time. Of course, this is only my personal opinion so that doesn’t mean much and certainly not that you’ll have to agree with me on this.

    As for *Pandion* and other monotypic genera, I feel the way to go is to consider them stem-based. In this case, that would probably mean ‘everything closer to *Pandion haliaaetus* than to, say, *Accipiter nisus*’. Here,where the family is also monotypic, we get the problem of there not being any fossil genera assigned to Pandionidae: *Pandion* becomes synonymous with Pandionidae. Had, say, the fossil Fayum pandionid been placed in a different genus (and time of origin alone would make that reasonable, I think), we might have defined Pandionidae as the node (*Pandion haliaaetus* + Fayum pandionid) instead.

    Link to this
  22. 22. John Harshman 3:50 pm 01/18/2013

    Can we all agree that without making extra assumptions a fossil can only give a youngest possible date for a node, and that this date must be at least slightly younger than the true date? And of course the assumption we would have to make to do any better than that would be that we have a good sample of avian taxa both in time and space. That’s just not something I’m willing to do. Most decent bird fossils come from a few Holarctic lagerstätten. If Aves (=Neornithes) and/or various clades within the group were limited to Gondwana for most of the Cretaceous and/or much of the Paleogene, I don’t think we would be very likely to have noticed them. Consider that oscine passerines, based on tree topology, spread out from Australasia in the neighborhood of three times, all of them fairly recent compared to the age of the oscine node. There just aren’t enough bird paleontologists, especially those looking for Cretaceous avians in Gondwana. And another problem is that many of the best sites might be under a mile or more of ice.

    I’m just saying that we should be wary of estimating time of divergence based on time of first appearance. Nothing weird here.

    Link to this
  23. 23. naishd 5:55 pm 01/18/2013

    As always, what John says is sensible and reasonable. I’m not wholly sure if comment 22 is a response to what I said above about hypothesised Mesozoic divergences within Galliformes, or to what BrianL says in comments 7 and 21 about using generic names for possible total-groups rather than crown-groups. However, with reference to comment 22 and those alleged Cretaceous galliform divergences, note that – while it’s obviously true that documented fossils will often or always be younger than actual divergence dates – the discrepancies involved here are pretty major: the genetic studies concerned are positing crown-galliforms as far back as 80 Ma (the middle of the Late Cretaceous), yet the fossils of such animals aren’t known until the late Oligocene*. This isn’t a “slightly younger” occurrence – it’s a discrepancy of about 55 Ma!

    As for the idea that the fossils that might fill in those ghost lineages are hiding in the Southern Hemisphere: sure, it’s possible, but it’s looking increasingly suspect given that stem-members of the respective groups are well represented in the north. If the crown-groups evolved in the south, wouldn’t we predict that the stem-taxa did too… yet, if that’s so, why are their fossils so well represented in the north?

    I don’t want to imply a literal reading of the fossil record since we know where that has led us in the past, but I now think that neornithine divergences that are posited to have occurred well prior to representation in the fossil record are most likely based on mistakes as goes calibration points, evolutionary rates and so on. All I say is: stay tuned, new stuff is coming…

    * I follow Mayr, Ksepka and other authors in regarding gallinuloidids, quercymegapodiids and so on as stem-galliforms.


    Link to this
  24. 24. naishd 5:59 pm 01/18/2013

    ps – yay, 23 comments :)


    Link to this
  25. 25. John Harshman 6:53 pm 01/18/2013

    If the crown-groups evolved in the south, wouldn’t we predict that the stem-taxa did too… yet, if that’s so, why are their fossils so well represented in the north?

    I would maintain that we wouldn’t predict such a thing. Consider that all non-avian theropods could be called stem-birds. Does finding a Tyrannosaurus in Wyoming predict that birds originated in Wyoming? The broader the clade, the more widely distributed we expect to find it, and distribution of the broader clade doesn’t necessarily say much about the distribution of the included clade. Anyway, I’m sticking with the poor quality and biased sampling of the fossil record as a potential explanation. Again, consider the oscines.

    Can’t wait to find out what you know about Vegavis, by the way.

    Link to this
  26. 26. naishd 7:22 pm 01/18/2013

    Yup, I will definitely agree that it’s a potential explanation. But it’s looking increasingly shaky.

    Oscines: yes, genetics indicates that this clade originated in Australasia before spreading north. But this is exactly consistent with the fossil record, since the oldest oscines known (late Oligocene lyrebirds and logrunners) are from Australia while the Oligocene passerines known from the north are stem-passerines!


    Link to this
  27. 27. John Harshman 7:54 pm 01/18/2013

    Note that the stem-passerines (zygodactylids?) do not suggest that passerines should also be found there if they had existed. Incidentally, my aforementioned stupid calibration of the Early Bird tree puts the oscine node at 35ma. Only 10 million years older than you would like?

    Link to this
  28. 28. naishd 8:05 pm 01/18/2013

    Yes, zygodactylids might be stem-passerines, but there are a number of other taxa I had in mind (Wieslochia and some unnamed French and German specimens). An Oligocene node for oscines isn’t (in my opinion) that bad – remember that the galliform discrepancy is more like 55 Ma. And it’s easy to come up with various excuses as to why the passerine record is so poor – do these plausibly apply to galliforms as well? Nope, mostly.


    Link to this
  29. 29. David Marjanović 8:51 am 01/19/2013

    Undoubtedly, you’re right in pointing out that apomorphy-based classifications should not be judged by cladistic principles

    …I said no such thing. Phylogenetic nomenclature doesn’t require cladistics, and node- and branch-based definitions are just as acceptable as apomorphy-based ones. Cladistics just means:

    1) (as used by morphologists) a cover term for the methods of the science of phylogenetics (now that it is in fact a science and not an art anymore);
    2) (as used by molecularists) plain & simple maximum parsimony.

    Can we all agree that without making extra assumptions a fossil can only give a youngest possible date for a node, and that this date must be at least slightly younger than the true date?

    Yes, but that doesn’t mean it’s never possible to get oldest possible ages for nodes from the fossil record. Sometimes the record is good enough that absence of evidence is evidence of absence. This is treated in Raaum (2005; no time to look up the complete citation now), in my first paper (with Michel Laurin, Syst. Biol., 2007), and in my next paper (again with M. L., Geodiversitas, scheduled to appear in March).

    If you only use minimum ages as calibrations, your results will be too old – maybe not by much, maybe only as much as to offset the fact that minimum ages tend to be too young, but maybe much more.

    BTW, has any additional research been done on the supposedly passeriform crumbs from the early Eocene of Australia (Murgon, I think; maybe Tingamarra)?

    Link to this
  30. 30. John Harshman 9:34 am 01/19/2013

    Sometimes the record is good enough that absence of evidence is evidence of absence.

    If we were talking about marine snails, I would totally agree. Birds, not so much, particularly in an understudied region. Got a PDF of that Geodiversitas paper? And is this Raaum 2005?: Raaum RL, Sterner KN, Noviello CM, Stewart CB, Disotell TR. 2005. Catarrhine primate divergence dates estimated from complete mitochondrial genomes: concordance with fossil and nuclear DNA evidence. Journal of Human Evolution 48:237-257.

    Link to this
  31. 31. David Marjanović 12:47 pm 01/20/2013

    Geodiversitas: no, we’re supposed to get the page proofs this month…
    Raaum et al.: yes, that’s it.

    Link to this
  32. 32. Max Blake 4:51 pm 01/20/2013

    Hydrarchos & Darren, to add a little more about the Welsh ‘serpents’, R. S. R. Fitter, in ‘The Ark in Our Midst: The Story of the Introduced Animals of Britain; Birds, Beasts, Reptiles, Amphibians, Fishes’ from 1959 gives an example (if my memory serves me correctly, I don’t have the book with me at the moment) of an incident when a naturalist went into an extremely isolated and rural part of Wales, and was told about the aforementioned flying serpents. Eventually, one was brought dead to him, and he identified it conclusively as a ring-necked pheasant (_Phasianus colchicus_). So in this case at least, the flying serpent was indeed a pheasant. So I would be pretty confident in saying that reports of flying serpents with bright scales/feathers in Wales can be explained best by rural communities being unfamiliar with a newly introduced species, the ring-necked pheasant.

    Link to this
  33. 33. Christopher Taylor 7:43 pm 01/20/2013

    @9: As well as the cold-resistance factors mentioned by Jerzy in comment 17, I think I remember reading somewhere that the green peafowl is natively a lot more difficult temperamentally than the blue peafowl, generally being a lot higher strung and less amoenable to tolerating the presence of humans. I’m happy to be corrected, though, if I’m remembering incorrectly.

    Link to this
  34. 34. naishd 5:43 am 01/21/2013

    Max (comment 32) – thanks for this, I had forgotten it.


    Link to this
  35. 35. Zoovolunteer 1:35 pm 01/21/2013

    As far as European legends being inspired by peacocks goes, considering how popular they were as ornaments and delicacies in Roman times, I suspect there were quite a few dotted around the empire (and afterwards). I believe there was at one time (maybe still is?) a feral population on one of the Greek islands as well, not to mention the way in the UK at-liberty birds have a habit of straying well away from their supposed place of residence. Plenty of opportunity there for them to enter mythology.

    Link to this
  36. 36. CS Shelton 5:09 am 01/29/2013

    BTW, as an artist and fuzzy dinosaur fan, one thing jumped out at me in this article – Snowy Peacocks! A Bulgarian peacock adapted for a cold environment. I’ve seen lots of images and video of pure albino peacocks and they’re cool, but I don’t imagine a healthy wild animal would look anything like that. I’m sure it would have notes of other tones in there, maybe change plumage during the seasons… Still, very exciting to imagine!

    It seems safe to assume it has no living descendants, so that opens the possibility of some very divergent soft tissue features. Did they have fuzzy feet? Were their trains extremely different, if they had them at all? The way color is divided on living peacocks into bold distinct segments makes me imagine something with strong large areas of white, and cleanly demarcated bits with a bark-like or rocky appearance. No iridescence. Or maybe concealed shiny bits, displayed for sexy dances. Whatcha think?

    Link to this

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

More from Scientific American


Get All-Access Digital + Print >


Email this Article