Thanks to Tet Zoo, I sometimes receive books to review, and earlier on in the year I was fortunate enough to receive a copy of Lewis Smith's Why the Lion Grew Its Mane (Papadakis, 2008). Smith is a science reporter at The Times and in this book, billed as presenting 'a miscellany of recent scientific discoveries from astronomy to zoology', he takes us on a tour of some of the newest, neatest science. Despite its title, Why the Lion Grew Its Mane isn't just about animals, but also includes sections on cutting-edge technology, astronomy, genetics and psychology. I don't know enough about any of those subjects to comment on them, so what you're going to get here is a review of only some of the book - namely, of course, the parts on tetrapods.
Basically what we have here is a hand-picked selection of newish science stories that are cool, involving charismatic or neat subjects. This means, I suppose, that coverage is biased towards dinosaurs, big cats, meerkats and so on, and if you keep an eye on new and breaking news, many of these discoveries will already be known to you: they're exactly the sort of things that get picked up by National Geographic, BBC science news, and the broadsheet newspapers. Think parthenogenesis in Komodo dragons, dinosaurs found in burrows, the Spanish mega-sauropod Turiasaurus, spear manufacture in chimpanzees, orangutan bipedality, Tiktaalik, the incredible tongue of the Tube-lipped nectar bat Anoura fistulata, evolution in action among Darwin's finches and Bahaman anoles, and the development of new songs in urban passerines. Here they are - together - bound up in a glossy, very attractive, large format, 'coffee table' book. Excellent photos appear throughout and the book really is a joy to look at. The volume is in fact so attractive that I tried hard not to be nasty about it, but read on.
The book takes its name from the section on baldness, virility, and mane size in lions Panthera leo. Lions are variable in mane size and extent, with males of some populations lacking manes entirely, and why lions have manes, and why manes are so variable, remain two of most debated questions in felid biology. Following research by Julian Kerbis Peterhans and colleagues, Smith states that fully-maned lions may in fact be 'beyond their prime' and that individuals with less-developed manes may be more attractive to females. He also notes that mane extent varies with climate, with lions from the hottest, most humid places having the thinnest manes. This is - based on what's been published - partly right, but it's also partly very wrong.
I presume that the research discussed by Smith is that eventually published by Gnoske, Celesia and Kerbis Peterhans (Gnoske et al. (2006) [as discussed below, sources are not cited in the book], but what's confusing is that Gnoske et al. (2006) didn't really say what Smith ascribes to Kerbis Peterhans. Concentrating on the (sometimes maneless) male lions of the Tsavo region [one shown here, from wikipedia], Gnoske et al. (2006) showed how mane development is not fully associated with sexual maturity (in other words, that mane development is extensively delayed in some populations), and that 'climatic adaptations to the relatively hot and humid climate of GTE [greater Tsavo ecosystem] has resulted in changes that may slow down and delay the onset of mane development and growth. The slower development of the mane and the smaller mane surface should, in theory, improve the potential for maintaining efficient thermoregulation' (p. 558) [on thermoregulation, see the thermographic image below, from wikipedia. The mane clearly has an insulative effect]. Kays & Patterson (2002) had previously argued that the sometimes maneless lions of the Tsavo region owed their sparse or absent manes to local environmental conditions (like the area's abundant thorn bush and hot, dry climate), Patterson (2004) proposed that particularly high levels of testosterone (itself associated with a social system where a pride contains but a single male, rather than a coalition) resulted in the reduced/absent manes of the Tsavo region, and Patterson et al. (2006) later showed how up to 50% of the variation seen in mane length and density was correlated with climate.
Why manes are so variable is clearly a complex subject and there are several competing views; you might have recognised that none of this really tells you 'Why the lion grew its mane'. I recognise that - given the short, snappy text and need to summarise - Smith was not able to indulge in complex, detailed discussions of the subjects he covered. Nevertheless, as demonstrated by the section on lions, I was left throughout the book with the impression that Smith had based his text on recent reports alone, and as a result his text merely presents one, recently mooted opinion on any given subject. Furthermore, his text often seems naïve in that it doesn't seem informed by the work that has gone before.
One section that will perhaps be of greatest interest to readers here is that on new species, and by now I suppose that most of the stories covered in this section will be highly familiar: Smith looks at the resurrection of the Sundaland clouded leopard Neofelis diardi (or diardii) (covered on Tet Zoo here), at the discovery of the Kipunji Rungwecebus kipunji (covered on Tet Zoo ver 1 here) and at the naming of the Cypriot mouse Mus cypriacus. Ah, the Cypriot mouse. Smith doesn't really repeat that most-disliked of statements (you know, the one claiming that the mouse was 'Europe's first new mammal in 100 years'), but he does at least allude to this idea, stating that discoveries of new mammals 'almost always [occur] in remote regions of the world such as South American rainforests', and 'Europe has been well trawled by naturalists over the centuries, making it hard to believe than anything as big as a mouse remained undiscovered'. Long-time readers of Tet Zoo will recall my mild annoyance at claims such as these when they were first mooted in the popular press back in 2006. In reality, even boring old, well-explored, no-new-species-to-reveal Europe yielded 32 new mammal species between 1906 and 2006.
Smith also writes about DNA barcoding of North American birds, and I was particularly interested in the discussion of white-headed gull taxonomy as a two-page photo-spread declares in giant writing that 'Over the centuries, eight gulls had been identified as separate species. Yet when they were assessed, their genes were found to be virtually the same' (pp. 96-97). Smith also says that 'one type of gull ... had been categorised as eight different species' (p. 95) and that these eight 'should all be classified as the same bird or at the very least only as subspecies, according to their DNA' (p. 95). The eight species concerned are the California gull Larus californicus, Herring gull L. argentatus [photo below by Neil Phillips], Thayer's gull L. thayeri, Iceland gull L. glaucoides, Lesser black-backed gull L. fuscus, Western gull L. occidentalis, Glacous-winged winged L. glaucescens and Glaucous gull L. hyperboreus. Hold on - who says that they should be regarded as the same species?
We all know that the taxonomy and phylogeny of the white-headed gulls is very complex and very confusing - there are multiple populations that grade into one another [though forget the 'ring species' concept*] and multiple views on whether populations should be regarded as subspecies or species - but I was not aware of any recent article which had promoted such a massive act of lumping. If anything, ornithologists have taken to splitting up the white-headed gulls even further: as recently as 2007 it was being recommended that, based on mtDNA data (Crochet et al. 2002, Olsen & Larsson 2004, Pons et al. 2005), certain 'subspecies' of L. argentatus should be elevated back to specific level (namely L. smithsonianus, originally named as a distinct species in 1862 but later demoted to a Herring gull subspecies, and L. vegae, named in 1887 and also later demoted to a subspecies of L. argentatus).
* I wrote about this at Tet Zoo ver 1 in 2006 [here] after being surprised at seeing Richard Dawkins' promoting the ring-species model in The Ancestor's Tale.
It seems that Smith had one particular piece of research to hand: Kerr et al.'s 2007 paper 'Comprehensive DNA barcode coverage of North American birds' (Kerr et al. 2007). Does this paper really conclude that eight morphologically distinct white-headed gulls should be sunk into one? No, it does not. What it does say is that the DNA barcodes of the eight species overlap somewhat, as do those of 17 other sets of North American bird species, and that there are three possible explanations for this. (1) That the species have only diverged very recently and have yet to accumulate sequences differences within cytochrome c oxidase 1, the mitochondrial gene looked at in the study (the barcode study was based on data from a single gene); (2) that the species are doing a lot of hybridising; or (3) that they really are part of a single species. I therefore feel that Smith's emphasis on the 'all those gulls are virtually the same' line to be misleading. Nobody doubts that all these gulls are very similar, but the study that Smith discusses did not come along and present some magic new perspective on this situation. In fact, if anything, Kerr et al. (2007) drew attention to the fact that their data indicated the presence of multiple cryptic species within the avifauna of North America.
Smith also includes a piece on the Kayan Mentarang animal: that peculiar, long-tailed, reddish mammal photographed at a Bornean camera-trap in 2003 (not 2005, as Smith says). By 2006 it was thought by many that the animal was not the new viverrid or mystery pseudo-lemur that some had thought but, in fact, a giant flying squirrel. I wrote about this at Tet Zoo ver 1 (2006) here and here, and Meijaard et al. (2006) published on it in Mammal Review. Given that Smith's book was published in 2007, and probably mostly written in that year (the acknowledgements are dated '2007'), it is surprising that he hadn't caught up on any of this.
A major gripe I have is that there is no bibliography whatsoever. I'm not entirely stupid: I know full well that publishers often don't want authors citing references and listing papers and articles at the back of a book. But this book has nothing at all; at the very least it could have included tiny (8-point or something) 'source notes' or such on each page (there is plenty enough space on the pages for this). A list of journal titles is given in the acknowledgements, but that's because Smith is acknowledging help, not citing sources. There is also, sin of sins, no index.
Anyway, enough complaining; this is a lavishly illustrated book that, despite my dissatisfaction, does a great job of bringing the wonder of science to the masses. Smith's discussions of the discoveries he writes about are short, succinct and only really tell the highlights. That might prove annoying when you want to know more, but the whole point is that that's not what this book is about: as it says on the back cover, Why the Lion Grew Its Mane is 'entertaining - a colourful tour through the world of cutting edge science'. If it encourages any of its readers to go deeper - if it ignites the spark that makes them want to know more about any of the subjects it covers, then it's done a great job and is a good ambassador for science, for knowledge, and for the wonder of discovery.
Smith, L. 2007. Why the Lion Grew Its Mane. Papadakis (London), pp. 287. ISBN 978-1901092837. UK price £20. See www.papadakis.net.
Refs - -
Crochet, P. A., Chen, J. Z., Pons, J. M., Lebreton, J. D., Hebert, P. D. N. & Bonhomme, F. 2003. Genetic differentiation at nuclear and mitochondrial loci among large white-headed gulls: sex biased interspecific gene flow? Evolution 57, 2865-2878.
Gnoske, T. P., Celesia, G. G. & Kerbis Peterhans, J. C. 2006. Dissociation between mane development and sexual maturity in lions (Panthera leo): solution to the Tsavo riddle? Journal of Zoology 270, 551-560.
Kays, R. W. & Patterson, B. D. 2002. Mane variation in African lions and its social correlates. Canadian Journal of Zoology 80, 471-478.
Kerr, K. C. R., Stoeckle, M. Y., Dove, C. J., Weigt, L. A., Francis, C. M. & Hebert, P. D. N. 2007. Comprehensive DNA barcode coverage of North American birds. Molecular Ecology Notes doi: 10.1111/j.1471-8286.2006.01670.x
Meijaard, E., Kitchener, A. C. & Smeenk, C. 2006. 'New Bornean carnivore' is most likely a little known flying squirrel. Mammal Review 36, 318-324.
Olsen, K. M. & Larsson, H. 2004. Gulls of North America, Europe, and Asia. Princeton University Press, Princeton.
Patterson, B. D. 2004. The Lions of Tsavo: Exploring the Legacy of Africa's Notorious Man-Eaters. McGraw-Hill, New York.
- ., Kays, R. W., Kasiki, S. M. & Sebestyen, V. M. 2006. Developmental effects of climate on the lion's mane (Panthera leo). Journal of Mammalogy 87, 193-200.
Pons, J.-M., Hassanin, A., & Crochet, P.-A. 2005. Phylogenetic relationships within the Laridae (Charadriiformes: Aves) inferred from mitochondrial markers. Molecular Phylogenetics and Evolution 37, 686-699.
- Log in to post comments
The relationships within the white-headed gull group are very complicated. Rather than lumping them all as one species the question is simply how far to go in splitting a group that appears to have diverged very recently. Most ornithological sources now treat all the forms you list as valid separate species (though separating some of them in the field is a bit of a nightmare - try telling apart adult Larus smithsonianus and Larus argentatus - it basically takes a DNA test since they are only visually distinctive in first year plummage). There are still some that are debated, for example the relationships of Steppe Gull Larus (cachinnans) barabensis and Mongolian Gull Larus (cachinnans) mongolicus to Caspian Gull Larus cachinnans. I also am inclined to think that Baltic Gull (Larus fuscus fuscus) is actually a separate species from Black-headed Gull (Larus fuscus graellsii and Larus fuscus intermedius) - contrary to what has been stated they are not known to hybridise (to be fair their breeding ranges, though relatively close do not actually come into contact), are visibly distinct, have distinctive 'jizz', and have very different migration routes and wintering areas - though most sources still treat them as conspecific.
I am suprised Dawkins still referred in 2004 to them as an example of a ring-species - that was never well supported, and as far as I am aware was killed-dead well before that as a gross over simplification of a very complicated set of relationships. Having said that I wasn't at all impressed by The Ancestor's Tale.
Hi Mark. Whoops, you accidentally called the Lesser black-backed 'Black-headed' (and, for those who care, black-headed gulls are no longer part of Larus but represent the separate genus Chroicocephalus. Cue chorus of disdain).
The ring species model was contested in...
Liebers, D., de Knijff, P. & Helbig, A. J. 2004. The herring gull complex is not a ring species. Proceedings of the Royal Society of London B 271, 893-901.
And for the Tet Zoo take on this go here on ver 1.
I regret to say that I was bored to tears by The Ancestor's Tale and it still sits, unfinished and unloved, on the shelf. Given its 'time deep' approach I was disappointed that, time and time and time again, he chose to write about the most boring things. And the artwork is bloody awful.
I too have disputed the ring species claim for the herring gull, and found the Ancestor's Tale deadly dull. But you do it so much better.
Not that I knew anything about this particular situation, but I like it anyway. Neontologists make their genera way too big to keep any oversight. I'm waiting for the utter busting of Varanus into twenty genera ;-)
On the subject of gulls if they ever did get lumped together I think a lot of birders would be mighty pissed that if they lost 7 species off their life list, especially if they had made a special trip for a rare migrant that would then be classed as a common herring gull!
David said...
In general I agree. Recent phylogenies of such things as gulls and terns have found that the huge, multi-species genera deserve to be split up, in part because highly distinct, 'traditionally recognised' 'genera' fall smack in the middle of the 'traditionally recognised' multi-species genera. Most bird books of the last 40 years have recognised Rissa, Pagophila, Rhodostethia, Xema and Creagrus as distinct from Larus. Now, Ichthyaetus, Chroicocephalus, Leucophaeus and Hydrocoloeus have been reinstated.
As for Varanus, the word on the street is that it's gonna stay as it is... some might say because of 'cultural inertia'.
Darren:
Are you sending Dawkins' publishers the message that they better not ask you to review his forthcoming books...?
First: you all stop lying about any discovery. Gulls, mice and clouded leopards were known for decades. In fact, most/all gulls and clouded leopard WERE DESCRIBED AS NEW SPECIES WHEN DISCOVERED, DECADES AGO. Since then they were sunk into subspecies and re-elevated again. The only claim of fame is: 1) finding some additional DNA and behavioral differences, 2) returning to "splitting" fashion which was present in 19. century zoology 3) big ego which pays no attention to researches from long ago, and proudly claims the same.
About barcodes: its another pretentious heap of rubbish. Lack of generic differences between large gulls was discovered long before by several DNA techniques, the same with lack of differences between other forms/ within-species difdereces of others.
About lesser black-backed gulls: current wisdom is that fuscus and intermedius/graellsii have full overlap in morphological differences (diagnosable only if ringed on breeding grounds) and no genetic differences. So certainly hybrdids are difficult to find... ;)
Jerzy writes...
Yes, that is exactly the point made above (and elsewhere). I don't see anyone claiming to have found 'new' species, merely to have resurrected old ones.
(OT) The aye-aye lemur taps wood with it's long clawed skinny middle finger, and probes for insect larvae, to as deep as 2 cm into the wood of Madagascar trees, and depends upon its finely tuned echolocation ability. Dolphins and bats use clicking/chirping to echolocate prey. Do woodpeckers use their continuous knocking to echolocate prey? Are their auditory equipment highly derived from typical Avian ears, or very much the same? I seem to recall another animal which uses a tapping and probing method similar to the aye-aye, (perhaps here at Tet Zoo?), but I've lost track of it.
The title of the book is a perfect example of lamarckian language (I think I spelled it right this time), and should have been "how the lion made use of it's mane", or "how natural selection shaped the lions' mane". Like I said, just a pet peeve.
That's no squirrel, that's a miniature brachiosauroid! Even more amazing, it has phosphorescent, er, coprolites-to-be.
DDeden:
hmm, by any chance was it the new caledonian crow? or were you thinking 'mammal'?
And what use is a book with no bio or index?
They are often the best bits.
Educate the ignorant! What's wrong with "the ring species concept"? Does it just not fit gulls? (E.g. because they are too mobile or something?) Or is there something wrong with the notion in general?
I notice Wikipedia's "Ring_species" topic still promotes a gully ring despite (somebody) citing Liebers "... gull complex is not a ring species". I also notice it leaves out a seemingly plausible topology that is actually ring-shaped with no endpoints, which might reasonably seem obtainable in, e.g., snails around Australia, or, I don't know, rosemary around the Mediterranean, or nematodes around Ireland.
To get an actual ring qua ring, I suppose one would need original propagation and establishment fast enough to have near uniformity at the outset, followed by local adaptation and then successful competition at adaptation centers against later interlopers and hybrids, and slow enough propagation to keep hybridization from diluting local adaptations. Has anything like this been observed?
If the ring notion isn't total rot, shouldn't the Wikipedia article be made refer to a valid example? It only just mentions two other examples, salamanders in California and warblers around the Himalayas.
Choirocephalus...
Split of Larus gulls into several genera is actually not justified by cladistics or nothing. The group (after deleting Xema etx) was monophyletic. it is nothing but a fashion.
The problem is that black-headed gull, like many other popular animals, is name used not just by a group of scientist, but much general people (e.g. toxicologists and conservationists) who don't follow the trends. So the new name created chaos.
Even worse, other ornithological regions still use Larus.
Darren Naish Writes:
I regret to say that I was bored to tears by The Ancestor's Tale and it still sits, unfinished and unloved, on the shelf. Given its 'time deep' approach I was disappointed that, time and time and time again, he chose to write about the most boring things. And the artwork is bloody awful.
I'll have to admit, I liked reading "The Ancestor's Tale", although I agree about the artwork in the book. (Why do so many natural history books like those terrible computer generated images?) I was surprised that I liked this book as much as I did, I usually have a very hard time getting through any of Dawkins' books as a general rule. I read "The God Delusion" and although I agreed with most of what he was saying in the book, he was starting to get on my nerves towards the end.
I quite enjoyed the Ancestor's Tale, it helped me to get Phylogeny and cladistics a lot better. (I think!) It to a bit heavy going at the end with the bacteria and such though.
Sounds like a must-read book. A lot of sustainable problems we face are related to the choices we make to stay in our comfortzone. More thermographic images, mainly of animals, can be found at www.worldofwarmth.com.
DDeden the animal that you are thinking of is probably one of the members of the genus Dactylopsila, variosly known as banded possoms or trioks.
They have similar wiry fingers to aye-ayes and use them to winkle grubs out of rotten wood.
A good example of convergent evolution.
So THAT'S what "barcoding organisms" means: substitute cytochrome oxydase sequences for systematics? I think if I were a Marxist I would now be mumbling about "de-skilling"! (As well as about mystification.)
Damn, I did mix up heads and backs, how embarrassing :0 - I'm glad I added the scientific name so it was fairly clear what I was actually referring to.
The split of Larus is one I have mixed feelings about. While I tend to agree that the genera into which it is proposed it is split do seem to be natural groups, I think maybe the split goes too far.
Varanus on the other hand looks well overdue for a good splitting.
And I know they are not tetrapods (not even animals) but what the hell is going on with genera like Euphorbia. Huge numbers of species, of vastly different growth forms, surely that genus should be split.
Neil, don't even joke about lumping the white-headed Larus species, I have my life-list to think of and I put a fair bit of effort into ticking several of them. Though I have to admit that I am not completely convinced about some of the splits herring gull/ american hering gull and common gull/mew gull are two splits that seem to be accepted generally now, but which I still have some misgivings about. Other potential splits are not recognised - for example kumlien gull was briefly treated (at least by some) as separate - but now is as far as I can establish always treated as a subspecies of iceland gull (I don't feel too bad about that as the decision to lump seems right and I got to see a vagrant at docks not far from my home, so didn't travel far for that one). As an aside, there are still some gulls in this group which don't seem to fit into any of the recognised forms. I recall in February 1999 visiting the Canary Is, and seeing four specimens of a form that didn't correspond with any I have been able to find in any guide (and I have quite a selection). If it had been a single bird I would have concluded it was abherent or possibly a hybrid, but a group of them are harder to dismiss that way.
Darren, you surprise me that the paper 'The herring gull complex is not a ring species' was as late as 2004. I know that it was mostly in the late nineties that I was trying to get my head around identifying these, and I am sure that in the birding press there were articles on particularly the central asian forms, which made it clear that the real situation was very complicated, and that the ring species concept while it may or may not work in other cases, does not work for the white-headed gull complex. Of course those articles were not in peer-reviewed journals.
Sort of. It means imposing a phenetic species concept: everything within a certain percentage of COI sequence similarity to the holotype belongs to the same species.
Sequencing COI is certainly a good idea for uncovering cryptic diversity, for example, but really imposing the associated species concept wouldn't do any good.
One important difference between the zoological and the botanical code is that the former contains an explicit list of the allowed ranks between genus and species: subgenus and "group of species". Nothing else has any standing. The botanists are not so limited and have been freely using sections, subsections, and I don't know what else (within subgenera) for centuries.
(Of course, eliminating mandatory ranks altogether would eliminate this problem altogether, too.)
That would make a lot of sense, but the word you're looking for is aberrant, "erring away from the Platonist truth". I'll check if it's already in the Eggcorn Database. :-)
On gull phylogeny and taxonomy, Jerzy writes
I do see your point, but I dont think that adhering with tradition or maintaining simplicity for the sake of it should be our aim, given that new nomenclatural proposals such as the one in question (Pons et al. 2005) simply try to better reflect phylogeny. Pons et al. (2005) did find a major group of gulls to be monophyletic, and it included all of the traditional Larus species. But it also included Rhodostethia (recovered as the sister-taxon to the Little gull), and given the clear division of Larus sensu lato into obvious 'masked', 'band-tailed', 'hooded', 'black-headed' and 'white-headed' clades, the most informative and useful decision (certainly in my opinion) would be to label these as separate genera. Members of all of these clades are all behaviourally and morphologically distinct, so doing so better reflects larid diversity.
I confess I do not understand it at all when people (usually those with no interest in or knowledge of phylogeny, like amateur birdwatchers and pet keepers) complain about nomenclatural changes. Units like 'genera' reflect our current understanding of diversity and phylogeny, so it is unavoidable that names will constantly change and that works such as fieldguides will be constantly made out of date. Sorry, but that's the price we pay for using binomials that are not just labels, but also encapsulate a hypothesis about what the organism is. Blame Linnaeus!
Incidentally, I forgot that Saunder's gull Saundersilarus sandersi was also found by Pons et al. (2005) to be distinct enough for 'generic' rank.
On the 'ring species concept' as applied to white-headed gulls, Nathan wrote...
Mayr proposed that a series of interconnected populations existed west-to-east (starting in the Aralo-Caspian region), and that at the far eastern end of this series (North America), the end members had crossed the Atlantic. However, they were now so distinct from the 'end' members of a series of populations that had moved west from the area of origin that both 'end' members were incapable of reproducing, and hence were separate species. The problems are (1) that the supposed chain of west-to-east populations is no such thing, but instead consists of distantly related gulls, each of which have their own biogeographical histories, (2) the supposed chain of east-to-west gulls also do not represent a series of interconnected populations, but also consists of distinct species that have their own biogeographical histories, and (3) the idea that the 'end' members of the west-to-east series and the end members of east-to-west series met up in western Europe and then found themselves unable to interbreed is hypothetical and there is no evidence showing that this really happened. The ring species concept doesn't work for white-headed gulls, therefore, because the model assumed the existence of a single, variable 'super species', whereas in fact what we seem to have is a group of very similar species that have moved all over the place and did not migrate from a centre of origin in simple, stepping-stone pattern.
This does NOT negate the ring species concept however, and Californian salamanders and leaf warblers around the Tibetan plateau do demonstrate it. See Stebbins (1949), Wake et al. (1986, 1989), Moritz et al. (1992), Highton (1998), Alexandrino et al. (2005) and Wake (1997) on the salamanders and Irwin (2002) and Irwin et al. (2001a, b) on the warblers.
Refs - -
Alexandrino, J., Baird, S. J. E., Lawson, L., Macey, J. R., Moritz, C. & Wake, D. B. 2005. Strong selection against hybrids at a hybrid zone in the Ensatina ring species complex and its evolutionary implications. Evolution 59, 1334-1347.
Highton, R. 1998. Is Ensatina eschscholtzi a ring species? Herpetologica 54, 254-278.
Irwin, D. E. 2002. Phylogeographic breaks without geographic barriers to gene flow. Evolution 56, 2383-2394.
- ., Bensch, S. & Price, T. D. 2001a. Speciation in a ring. Nature 409, 333-337.
- ., Irwin, J. H. & Price, T. D. 2001b. Ring species as bridges between microevolution and speciation. Genetica 112-113, 223-243.
Moritz, C., Schneider, C. J. & Wake, D. B, 1992. Evolutionary relationships within the Ensatina eschscholtzi complex confirm the ring species interpretation. Systematic Zoology 41, 273-291.
Pons, J.-M., Hassanin, A., & Crochet, P.-A. 2005. Phylogenetic relationships within the Laridae (Charadriiformes: Aves) inferred from mitochondrial markers. Molecular Phylogenetics and Evolution 37, 686-699.
Stebbins, R. C. 1949. Speciation in salamanders of the plethodontid genus Ensatina. University of California Publications in Zoology 48, 377-526.
Wake, D. B. 1997. Incipient species formation in salamanders of the Ensatina complex. Proceedings of the National Academy of Sciences 94, 7761-7767.
- ., Yanev, K. P. & Frelow, M. M. 1989. Sympatry and hybridization in a "ring species": the plethodontid salamander Ensatina eschscholtzii. In Otte, D. & Endler, J. A. (eds) Speciation and Its Consequences. Sinauer Associates, Inc., Massachusetts, pp 134-157.
- ., Yanev, K. P. & Brown, C. W. 1986. Intraspecific sympatry in a "ring species", the plethodontid salamander Ensatina eschscholtzii in southern California. Evolution 40, 866-868.
"Units like 'genera' reflect our current understanding of diversity and phylogeny, so it is unavoidable that names will constantly change and that works such as fieldguides will be constantly made out of date."
Darren, you just admitted that scientific names lost the purpose why they were introduced - they are less constant identifier of a taxon than local name.
Which is true among gulls - when I search for black-headed gull, I type in search enging "black-headed gull" to avoid missing part of literature using "Larus" and part switched to Chroicocephalus.
BTW, Dutch taxonomic commitee from which Mr. Sanger migrated to British commitee has a record of splitting long-established units and back-pedalling. For several years they splitted cormorants Phalacrocorax and Shag was Schistocarbo aristotelis. Then they turned back. They split Lesser Black-backed Gull into three species, only to lumping them again and admit that no diagnostic characters of these forms exist AT ALL. Within 10 years they went from one species to three: Larus fuscus, graellsii and heuglinii and then to opinion that races are only identifable as birds ringed on breeding grounds. Call it bad science.
Why is it 'bad science' to change your recommendations when those recommendations are based on the conclusions of published academic research? Sangster (not Sanger) and colleagues have tried to keep up with changing conclusions (e.g., Knox et al. 2002, Parkin et al. 2003, Sangster et al. 2004a, b, 2007) and have not made ad hoc decisions as you imply. The limits of human knowledge demand that conclusions constantly change the more we learn.
And why is it so hard to keep up with the changes anyway? Are people really so dumb that they can't find these things out?
Refs - -
Knox, A. G., Collinson, M., Helbig, A. J., Parkin, D. T. & Sangster, G. 2002. Taxonomic recommendations for British birds. Ibis 144, 707-710.
Parkin, D. T., Collinson, M., Helbig, A. J., Knox, A. G. & Sangster, G. 2003. The taxonomic status of Carrion and Hooded crows. British Birds 96, 274-290.
Sangster, G., Collinson, J. M., Helbig, A. J., Knox, A. G. & Parkin, D. T. 2004a. Taxonomic recommendations for British birds: second report. Ibis 146, 153-157.
- ., Collinson, J. M., Helbig, A. J., Knox, A. G. & Parkin, D. T. 2004b. Taxonomic recommendations for British birds: third report. Ibis 147, 821-826.
- ., Collinson, J. M., Knox, A. G. & Parkin, D. T. & Svensson, L. 2007. Taxonomic recommendations for British birds: fourth report. Ibis 149, 853-857.
Mark Lees:
Ah, February 1999. That was still in the era before Mullarney et al.'s Collins Bird Guide, was it not?
Actually, that book's treatment of the 'herring gull'-complex made it slightly outdated already by the time it was published. (But it's still the best available bird guide for any part of the world.)
Sorry for name mistake.
Yes, it is bad science to propose big changes basing on fragmentary or preliminary DNA research, when it contradicts other published studies and before more comprehensive DNA study comes out. Even if research is peer-reviewed.
Also, it is nothing but pompousness, when you change a working system used for a century, because the new one is weakly better or not better at all.
Especially if you are a commitee overseeing nomenclature of names of common birds, supposed to be used for communication by thousands of scientists and tens of thousands of people related to environmental matters in various ways.
Jerzy (thanks for your continuing comments): while I can see the merits of your argument, I simply do not agree that we should maintain a century's worth of taxonomic stability simply because that is what we've always done. This is simply an appeal to authority, and for many groups it has definitely obfuscated our understanding of diversity.
Time and time again, 'conventional' taxonomies have been maintained simply because they have gone unchallenged, not because they are correct. As you know, the massive multi-species genera that we today regard as 'traditional' (e.g., Larus sensu lato, Felis sensu lato, Lacerta sensu lato, Rana sensu lato, Bufo sensu lato) are artefacts of a 20th century lumping phase which was even more lazy and subjective than the Victorian-era super-splitting that went on before. Indeed, there has been widespread complaint among students of the respective groups that the maintenance of these massive 'traditional' genera obscured diversity and that they should be split up for the sake of convenience. When the publication of new data (mostly from mtDNA) has led to the recommendation that 'old' generic names should be resurrected (or that new ones be coined for specific clades), it actually rarely comes as a surprise. And note that taxonomic recommendations are generally conservative (typically relying on evidence from more than one study), plus: they are only 'recommendations' anyway, not mandates.
While getting rid of genera would be nice, it seems to me that the chief trouble here is with species. A vernacular label refers (more or less unambiguously) to some life-form whether it's considered a species or subspecies (or variety, etc), but the scientific name has to be changed every time one changes one's mind on its (sub)specific status.
Darren:
(emphasis mine)
If a main reason for breaking up genera is "convenience", then how is splitting, in principle, any less subjective than lumping?
Yes. Lots of people foolishly believe that university libraries are the holiest of holies, accessible only to initiates, so they don't dare enter.
Add to this the fact that many university libraries are, to my surprise, indeed not public, and the fact that on the Internet almost all primary literature is behind a paywall, so I have to retract "dumb" immediately...
That's an important point: there is no official classification. Unlike phylogenetics, taxonomy is not a science (except for the application of species concepts); all classifications are equal. If you want to say Larus tridactylus rather than Rissa tridactyla, nobody can stop you (except maybe a journal editor, and they shouldn't). Every classification is right, and no classification is wrong.
Dartian: I said that there has been widespread complaint that they ('they' = the huge, multi-species genera) should be split up for convenience... but that doesn't mean that they were (rather, people just talked about it a lot); they weren't formally split up until testable phylogenies doing so were published. The lumping was entirely subjective (e.g., lots of gulls look alike, so all should be included in one massive genus); the splitting being promoted these days (for groups such as Larus sensu strico) is based on the discovery and recognition (via parsimony methods) of clades.
Darren,
European birds are intensively studied for over a century. So, perhaps simply nothing is left in taxonomy...
Some genera are big, because... they biologically are. There simply exists such a thing as rapid speciation making lots of forms. Example are these large gulls.
Unfortunately, these recommendations are often made on just one paper, and some papers are very weak. There is also trend in subjectivity. For example, gulls Larus argentatus, cachinnans and michahellis hybridise freely in some areas in C and E Europe. This was ignored. It is well established that Gyrfalcon and Saker Falcon (besides little DNA difference) freely hybridise in Altai in Asia. This is again, ignored.
Unfortunately, there is also a trend in society of treating these "recommendations" as Holy Grail. Much of it is misplaced need of order - we get one commitee of specialists to do thinking for us. Unfortunately, result is chaos.
Overall, the splitting trend in birds is powered by two things. One is wish of birdwatchers to get more "ticks". Somehow, subspecies doesn't make good tick. Second is decision made long ago by Birdlife Internationasl that Red List should include only full species. The reason, valid at that time, was that there are too many subspecies to care about every one. The result backfired - people (who are often both scientists and conservationists) started splitting subspecies into species to get their beloved creature more recognition.
Jerzy: it may be clear that we won't agree on this subject :) On genera, I would say that you're missing the point: this being that generic 'limits' are entirely subjective and that 'genera' should be divided according to what we find most useful in terms of conveying information about relatedness. I say that giving separate names to distinct clusters of gull species (= clades) is definitely more useful than having one name for all those c. 50 species. While all of these c. 50 do form a geologically recent radiation, the clades that are now being awarded generic status are as different from one another as are many other classically recognised 'genera'.
On species and subspecies, given that species grade into one another, groups of workers should decide among themselves as to which nomenclatural system best fits the evidence (you should imagine species as arbitrarily defined segments of lineages, not as rocks in space). Hybridisation is often irrelevant to 'species' status as even many distant and undeniably distinct species can and do interbreed in the wild. And while many 'subspecies' do appear to be subsets of a given species (in which case, a subspecific classification is appropriate), others are as distinct morphologically and genetically as are other populations traditionally recognised as 'species'. How 'distinct' a population has to be before it becomes a 'species' varies according to the group concerned.
Ultimately, a conservative adherance to 'traditional' genera and species ignores the fact that our taxonomies are ephemeral, both because new data will ALWAYS come along, and because we should devise the most useful ways of conveying information about relatedness.
Perhaps one of the reasons these arguments (discussions?) go on so long is that there are at least two different audiences which are expected to share the results. One group of people -- the more numerous by far -- are the laymen who are interested in identifying birds for their own personal enjoyment. They like to mark off different types of birds on big lists. Some of them like to tell each other about their observations ("Hey, Joe, I saw a black-backed gull just do the most amazing thing ...").
Another group of people -- a smaller group, and one which may not spend as much time in the field -- is most interested in conveying information about the degree to which different birds are related, which in turn says something about the manner in which they evolved.
In some alternate universe, these two groups might come up with two different ways of identifying birds. The birdwatchers could keep their lists over the centuries, making only a few celebrated additions when a completely new type of bird is discovered, and (alas) many deletions as birds disappear. No names would change. The other group could maintain an ever-changing set of names which would depend on the latest and greatest genetic research.
At the moment, we're stuck in a non-ideal world in which a single list of names is failing to satisfy both groups.
If I were king of the world, I'd tell the birdwatchers to keep their existing lists and just ignore the changes in the technical literature. That would lead the largest number of people being the most content. A few scientists who communicate with the public would occasionally have to say things like "The large whitish gulls which are commonly called black-backed gulls, but which are really a mixture of the following species ....." I think they could handle this inconvenience. Maybe I'm wrong.
Just my two cents:
I prefer the Taxonomic (traditional) nomen rather than cladistics simply because it makes it easier to see relationships. But in order to see those relationships correctly the data must be updated when errors are found. Assigning a clade simply is not enough, we need, at a minimum, an Order, Family and Genus to every species and hopefully can specify subspecies and tribes to clarify the picture. For example, calling an animal a basal theropod tells me nothing about the animal. It needs a family for comparison to later finds (a set of criteria to relate to).
Darren, we indeed agree to disagree.
"'genera' should be divided according to what we find most useful in terms of conveying information about relatedness."
Why relatedness? Why not most useful in terms of conveing information about species apperance and behaviour?
"While all of these c. 50 do form a geologically recent radiation, the clades that are now being awarded generic status are as different from one another as are many other classically recognised 'genera'."
"Hybridisation is often irrelevant to 'species' status"
Certainly, gulls are d*mn uniform...
But - wow! In this line of thinking, you must award species and genera status to races of dogs and cattle. No matter whether mastiffs, chihuahuas and dachshunds appeared recently and can hybridise (can they?). They are more different as fox from a wolf!
The same should apply to humans. Actually, Khoisan (Bushmen), Tasmanians and Australian Aborigines could diverge from the rest of us no later than some gull "species" diverged...
You confuse different issues here. Cladistics is the method of the science of phylogenetics: how to reconstruct a family tree. It has nothing to say about classification (the domain of taxonomy). Taxonomy may or may not start from a phylogenetic tree, which may or may not have been reconstructed using cladistics.
Untrue. It tells you a lot about what character states the animal has and lacks, what kind of ecological niche it has, how big it is, and, finally, that it is dead. :-)
That depends. If you use one of the two Biological Species Concepts, it is relevant, if not the only criterion.
That said, it appears that a very small and shrinking number of people uses these concepts, with morphological phenetic and phylogenetic concepts being more widespread...
Hi Jerzy. I think this might have to be the last contribution to this most interesting discussion...
On naming and dividing genera, Jerzy says "Why relatedness? Why not most useful in terms of conveing information about species apperance and behaviour?"
A: because everyone involved in systematics agrees on the idea that organisms should be grouped together according to phylogenetic hypotheses. Extreme examples: longclaws and meadowlarks, golden moles and marsupial mole, sugar glider and flying squirrel, poison-dart frogs and Madagascan mantellids, etc etc.
As for the idea that domestic dogs and human races are potentially distinct enough to be worthy of taxonomic recognition, I made the point earlier that there is not one rule that applies across all groups: I would argue that groups of workers should decide among themselves as to which nomenclatural system best fits the evidence for the group concerned (this has been called 'the frog's eye view vs the bird's eye view').
That's an interesting point Jerzy though one which is unlikely to gain much favour. I would assume that doglovers would be rather upset by the splitting of their breeds.
From the little I understand of anthropology (and human nature) I can only guess at the sh*tstorm that splitting homo sapiens would cause for obvious reasons - mainly that humans are completely unable to discuss race sensibly (perhaps understandable given human history!!).
It's a fair point though. A lot of species seem split on rather random features and your generic Aborigine doesn't bare a lot of similarity to your generic Swede. It has always interested me that pre-tertiary vertebrates seem to be named in a completely different way to mammals. I recently was reading Mammoths, Sabretooths and Hominids (http://www.amazon.com/Mammoths-Sabertooths-Hominids-Jordi-Agusti/dp/023…), one of the excellent books that Mauricio Anton illustrates on mammal eveolution, and it said that mammals (at least fairly recent ones) are typically named as lineages of ancestral species which seems extremely subjective to me.
On the general public's awareness of nomenclature I think that the majority of the public are both unaware and disinterested in this kind of thing and see changes by scientists to defined groups as being merely annoying. I recently was told by the leader of an important and long term regional environmental project that phylogenetics had no utility, usage or benefit over traditional linnaean classification. If he does not see the use of it then you can only imagine how most ordinary people see it and changes in classification in general. There is a fundamental misunderstanding of what genera and species represent and evolution in general in the public domain. I wish often that scientists in general would spend more time attempting to educate the world rather than bicker with their peers and hide all their work in obscure journals that are unavailable to almost everybody. I live in Auckland, a city of 1.5 million people, and the only place I can read journals is at the uni library where I can't take them out or get access to JSTOR. Useless. It's no wonder people fail to see the relevance of most work done.
It's time to freeze the genus-species names, as much as possible. That would mean a moratorium on generic splits, and not merging splits that are only historical. Far and away the most useful quality of a name is its stability.
Generic names offer a clue about phylogeny, but they can never do more than that, because they cannot all be right howsoever you tinker. Just give it up, let the field guides alone, and pursue phylogeny on terms that don't generate spurious conflict. There's plenty of real conflict to resolve without borrowing more.
The lion in that picture looks not just maneless but downright hairless!
David
Re: Untrue. It tells you a lot about what character states the animal has and lacks, what kind of ecological niche it has, how big it is, and, finally, that it is dead. :-)
Other than that it is dead I disagree. Because it is basal we can't really draw these conclusions. Did it have the habits of its ancestors or did it lead the theropods in new habits. Basal does not tell us it's size, the parent lineage does. If it does not fit into a known family then it should be the basis for a new family. And as for cladistics, I have seen animals that are not close taxon put into the same clade. It's sometimes nonsenical.
They're still more similar to each other than to your generic African. Skin color is six genes, hair shape I suppose one other... out of at least 18,000.
It's also different to find reproductively isolated human populations. There was one on Easter Island for 400 years, and maybe one on Tasmania for up to 10,000 years, but that's it.
Actually, you are arguing for abandoning genus names and making uninominal species names...
Well, we can't get very precise unless we know what the person in question meant by "basal", which does not have a definition, but that doesn't mean it doesn't tell us anything. Imagine something like Coelophysis, with a shorter head and neck and a slightly less reduced 4th finger but otherwise about the same size, and you have a basal theropod. No?
No, the combination of both does. More derived theropods evolved in all kinds of direction; something basal is expected to stay closer to the ancestral condition.
1. I get to tell what "fit" means. And so do you. And so does everyone else. "Family" does not have a definition.
2. What would all those redundant families (identical to a genus, and usually a species, and often even a single known individual) be good for?
What do you mean by "close"?
Also, remember that "clade" simply means "an ancestor and all its descendants". All known life forms a clade, for example. You can have clades within clades.
David Marjanovic:
"Redundant" families?!? Such as?
Phthinosuchidae is a synapsid family consisting solely of a partial skull from the Permian of Russia. The family exists solely because traditional taxonomy demands that the ranks of genus and family exists between the type species Phthinosuchus discors and the containing order Phthinosuchia (which, apart from this skull, contains a few jaw fragments and postcrania, all supposedly belonging to a single species of Phthinosaurus).
Andreas: I realized after I sent my comment that David probably meant fossil taxa, not recent. But the fossil record is woefully incomplete, as we all know. Why should missing information on extinct organisms be a reason to radically change the classification system of the modern ones?
P.S. Regarding your example: isn't there another species, Phthinosuchus borissiaki, in that genus (and thus family)?
Dartian: in unranked taxonomic schemes, any higher taxon that consists of a single taxon is redundant. So, Dulidae is redundant with respect to Dulus dominicus, and - to chose an extreme example - Cycliophora, Eucycliophora, Symbiida and Symbiidae are all redundant with respect to Symbion pandora.
Wikipedia does list a species Phthinosuchus borissiaki, but considering that the type species of of Phthinosaurus is P. borissiaki, I expect this is the result of either incompetent lumping (Phthinosaurus Efremov 1940 would have priority over Phthinosuchus Efremov 1954) or simple misreading.
Darren: What about the extinct ancestors of Dulus dominicus? It obviously must have had some.
And even if we never, ever find any close relative of Dulus dominicus, either living or fossil, I still don't see how the observation
'species X happens to have no close relative that we know of'
leads to the conclusion
'the traditional classification system of all organisms must be discarded!'
That's not what David said; he merely asked what redundant taxa are good for. About the only answer I've seen seriously argued is conformance with he traditional classification system - if you, like David, think that the traditional system needs to be reformed anyway, you're bound to find that answer unsatisfying.
Redundant = monotypic. Almost (!) required by the ICZN, but, well, redundant.
Not only, no.
If we do find any, we aren't forced to put it into the same family. Again: "family" is not defined.
And yet, there are people out there who believe they quantify biodiversity, in peer-reviewed papers, by counting families. Families are not countable.
I think monospecific genera are justified when: (1) the species does not fit comfortably into an extant genus; and (2) when the sister group is a genus. Same reasoning would apply to higher groups as well. I've been involved in describing two monospecific genera justified, in my mind, by the two criteria. That's my story and I am sticking to it!
Ah, Larus gulls. I'm thoroughly confused by the changing classifications of them - when i was a child in the Young Ornithologists' Club, the large gulls in the UK were Herring gulls (L. argentatus), Lesser black-backed gulls (L. fuscus) and Greater black-backed gulls (L. marinus). Now i have no idea how many species there are...
In Birmingham, there seem to be almost a continuous range between the palest-grey-backed, pink-legged obvious argentatus and the nearly-jet-black-backed, yellow/orange-legged obvious fuscus (they're definitely not big enough for marinus, in fact there seems to be a rough inverse correlation between size and wing pigment). If there really are several discreet populations, then i reckon there must be at least 3 intermediate "colour phases" between those 2 extremes.
The commonest seem to be the gulls one shade lighter than the "true black-backs" (perhaps. L. (fuscus) graellsi?) and those one shade darker than the pink-legged "true herring gulls" (which have pale yellow legs, and could be michahellis or cachinnans? not really sure...) There's an intermediate (sort of slate-grey) between those, tho...
As they're all out of their "natural" habitat in an inland city anyway, i'm guessing that the resulting alterations in breeding behaviour might make them less reproductively distinct and more likely to hybridise with each other than in less human-influenced ecosystems where they have their own niches (a bit like ducks and geese on human-made lakes/ponds)?
Anyway, if anyone does know if there's a consensus on how many species of large gulls there are in England, i'd be interested...
Re:
1. I get to tell what "fit" means. And so do you. And so does everyone else. "Family" does not have a definition.
2. What would all those redundant families (identical to a genus, and usually a species, and often even a single known individual) be good for?
Lets say we have a genus description of a good specimen. Then we find another that fits the genus mostly but is different enough to warrant another genus. The family description would have the commonalities of both. What it is good for is to help see the relationships better.
On cladistics, I don't disagree but feel that the description should include more than just a clade. I think that some of the younger scientists are getting a little out of touch with tradition, thats all.
Uninominal species names? Effectively, yes. Germans might find that easy to deal with, but the rest of us need the occasional word break. A more or less arbitrary (preferably less, but not enough so to agonize over) genus assignment makes the vocabulary more practical.
Remember what we-all are paying you folks for. We don't have time to follow the literature and keep up with changes; that's your job. Just give us names that are as stable as can possibly be sustained. Species may frequently be synonymized (without much consequence) and occasionally split (with a little more), but the churn that comes with phylogenic controversy is destructive.
This posting is not meant as a joke.
Here is were you go for common and scientific names of North American fishes. http://www.afsbooks.org/x51029xm.html
These common names are much more stable than scientific names. It used to be that a "shiner" belonged to the genus Nortopis (Well, the golden shiner is not, and was never, a Notropis.) Today many shiners are no longer Notropis species, but their common name persists. I haven't counted them up, but I would not be surprised if most shiners today are not Notropis species. For example: blacktail shiner, common shiner, redfin shiner, etc. Fisheries biologists tend to use common names while ichthyologists tend to use scientific names.
A couple of comments; first, a more specific, then some more general ones.
It just rubbed me the wrong way when David said
(To ask what something is "good for" sounds like the kind of teleological language that should never be used in a serious biological context, but never mind that for now.)
The way I read it David was asking a leading, value-laden question. To say that monotypic genera are "superfluous" is not a statement of fact or even generally agreed consensus, it's a personal opinion (and a very subjective one at that). Who exactly is this Almighty Authority who, in his/her infinite wisdom, has declared certain taxa superfluous, and that genera (or families, or orders, or whatever) must have a certain minimum number of members? And why should I, or anyone else for that matter, just flatly accept that opinion? Original references, please!
And since there is no need for beating around the bush anymore, let me just mention by name the subject that this thread has boiled down to - the PhyloCode - and proceed to my more general comments.
I have to be counted against those who oppose the PhyloCode (or at least its current incarnations). I do not oppose phylogenetic systematics, quite the contrary! It's the phylogenetic nomenclature I have a problem with. I am well aware that the PhyloCode has some very enthusiastic supporters among the paleontological community, and that many arguments have been made in its favour. But I'm yet to be convinced about the PhyloCode's practical utility.
The bottom line is this: What PhyloCode supporters are suggesting, even in the most moderate versions, would necessarily imply a tremendous upheaval of the existing names of organisms. The most extreme suggestions - abolishment of the binomen - sound like a taxonomist's nightmare.
Anyone who is really serious about replacing the existing Linnean nomenclature needs to come up with a significantly better alternative. And significantly better means that it must be superior in both theory and practise. You may talk about the importance of 'defining taxa' until you are blue in face, but at the end of day, that doesn't matter a whit unless your system is also user-friendly. Perhaps you don't give a crap about the opinions of non-systematists or non-paleontologists, but those others are very much in the majority and you ignore them at your own peril. Discarding the Linnean system would affect all biologists, everywhere, regardless if they are geneticists, ecologists, conservationists, microbiologists, physiologists, ethologists, biogeographers, embryologists, anatomists, etc. Whatever alternative to the traditional classification you are going to offer has to be so utilitarian that both Carl the Crafoord Prize Winner and Joe the Biology Teacher can see its merits. It's simply not enough that the PhyloCode is more logical or has a sounder theoretical base (and many would argue against that, too). If the PhyloCode is too cumbersome and unpractical to use, people will just vote with their feet and continue to use the Linnean system.
To illustrate what I mean, I'll use Darren's example organism. Under the Linnean nomenclatural system, you can say this (sentence taken almost verbatim from the Palmchat's wikipedia page):
"The palmchat, Dulus dominicus, is a passerine bird, the only species in the genus Dulus and the family Dulidae."
This simple sentence manages to convey quite a lot of taxonomic information for anyone who knows anything at all about birds. But how would you re-phrase that exact sentence by using phylogenetic nomenclature? Could you say it more simply and more succinctly? (These are not rhetorical questions; I genuinely would like to know. Every PhyloCode resource I've so far seen is awfully short on these kinds of practical, real-life applications.)
I did not write this to change anyone's mind. I intend this more as a gentle reminder of the fact that the eventual victory of the PhyloCode is by no means a foregone conclusion. The Linnean system has survived for 250 years precisely because of its utility and its 'common man-appeal'. If the PhyloCode is going to prevail, at the very minimum it needs to match that utility, and preferably surpass it.
Dartian: a couple of very brief comments you might like to take on board (not a proper response; I've spent too long responding to comments here already methinks)...
Even if the PhyloCode didn't exist, I and many others like me would like to see Linnean ranks abolished entirely. One good reason: they simply ARE NOT FAIR! They create the impression that groups of equal rank are somehow equal in diversity and 'importance', whereas in fact they are of course nothing more than historical artefacts.
Look at avian 'Orders'. Let's say Coliiformes. Now look at one of two conventional dinosaur 'Orders', Ornithischia. It is clear that ornithischians contain an order of magnitude more diversity than colies, yet - because of the Linnean system - people actually believe that these groups are somehow equivalent. In an effort to maintain the Linnean system, yet at the same time acknowledge the many subdivisions you have to introduce in order to usefully divide an 'Order' like Ornithischia, people initially used a whole slew of new subordinate ranks: Sereno (1986) had Parvorder, Nanorder, Hyoorder, Minorder, Infraorder, Microorder, Gigafamily, Megafamily, Grandfamily, and Hyperfamily (in addition to Superorder, Superfamily, Family, Subfamily etc). You might say: yeah, but everyone knows these ranks are just for convenience, and are not real biological entities. But, fact is, so long as ranks are used, some people actually believe that they are real, and that one ornithischian 'Family' is actually, really, biologically equivalent (in term of diversity) to one coliiform 'Family'.
So, please: unranked classifications.
Regarding the affinities of Dulus, in an unranked phylogenetic system this taxon would have to be described with reference to other clades (e.g., Dulus is a member of the passerine clade Passerida and is currently inferred to be the sister-taxon to Bombycillidae within the Muscicapoidea*). You might not think that's 'better' than an example using Linnean ranks (it's certainly not more succinct), but it tells us a lot more about the organism's affinities.
* I'm not necessarily saying that this is currently the favoured view: Johansson et al. had a new paper out this year that I haven't seen yet.
Darren, thanks for the reply. I don't want to drag this discussion on either, but I would still want to make one more comment.
You approach this situation from a paleontologist's point of view. But the vast majority of biologists are neontologists, who - sad to say - often don't really care about fossils. They do not see why the names of modern bird groups (for the purposes of a field guide, or a museum exhibit, or something such) would have to change just because Mesozoic dinosaurs were very diverse.
I see a potentially great risk in this, which is one more reason why I don't support the adoption of the PhyloCode. It is this: if paleontologists persist in using the PhyloCode, and neontologists in using the Linnean system, there will be a de facto separate paleontological nomenclature. And that nomenclature would be even more different from the already existing separate zoological, botanical, and bacteriological nomenclatures than these three are among themselves. If that happens, paleontologists might end up being 'kicked out' from the biological community (figuratively speaking) by the neontologists. I.e., there would be virtually no interdisciplinary communication or cooperation. And that kind of situation would harm paleontology much more than neontology. I do not want that to happen. Cutting the nomenclatural links between paleontology and neontology is too high a price to pay.
Look, at some level I sympathise with the PhyloCode project. But reality is what it is. Paleontologists are not, and likely never will be, in any position to force their views on nomenclature upon the rest of the biological community. And 'fairness' does not enter the equation.
Thanks for your thoughts (and I want to wrap up here too). But, woah woah woah, this is not palaeontology vs neontology (the PhyloCode's most vociferous proponents are best known for their work on extant squamates) - maybe I created the wrong impression by using a fossil group. Well, even with extant clades it's still obvious that ranks are downright misleading.
What if you discover a species that is the sister-group to a clade composed of two phyla? You'll likely end up creating a superphylum, a phylum, a class, an order, a family, and a genus for just that one species. Does that help anyone? Is it good for anything?
No, we don't find that -- because it's not a fact. We feel it. We subjectively decide that it's "different enough". Neither "genus" nor "different enough" have a definition.
And then someone "finds" that they are after all "similar enough" to be in the same genus. And then we have two competing classifications in the literature, of which neither is wrong. That happens all the time.
What do you mean? What do you mean by "description"? ~:-|
Cladistics was invented in 1950
Relax, the PhyloCode will not introduce uninominal species names. Read what it'll do instead.
I am not talking about science here. I am talking about nomenclature, which is a set of conventions, not science.
Again: nomenclature, not science. And not taxa, but names are superfluous.
As Darren already said, it's not a paleo- vs neontology thing. The botanist Phil Cantino, first author of the PhyloCode, is a neontologist, and so are all other prominent botanists in the International Society for Phylogenetic Nomenclature (like Nico Cellinese, who has arranged the Peabody Museum herbarium according to a phylogenetic tree -- with labels provided by phylogenetic nomenclature -- rather than as an alphabetical list of families, or Brent Mishler or Dick Olmstead). There are also people who work on animal taxa with a poor or no fossil record, like sponges, flatworms, annelids. On the other hand, I'm not aware of any trilobite or conodont workers who use phylogenetic nomenclature, and the Cenozoic mammal people have AFAIK been rather slow to adopt it.
Well, except for some people in Russia and China, everyone (!) who works on Mesozoic dinosaurs nowadays uses phylogenetic nomenclature, so it can't be that bad... :-)
You are two years behind the times. Please check out the link I give above.
That's my experience... I completely agree with your point about practical utility; and it is my experience that phylogenetic nomenclature is less cumbersome and more user-friendly than rank-based nomenclature, and leads to a lot more stability (because splitting and lumping, except at the species level, are impossible in phylogenetic nomenclature).
Taxa are not defined. Their names are defined. That's an important distinction.
"The palmchat, Dulus dominicus, is a passerine bird without close relatives; its closest known relatives are the [whatever]."
Two differences:
- The traditional wording also conveys the "without close relatives" part, but in addition it pretends to quantify it. If it would really quantify it (think Phylogenetic Diversity Index), that would be great, but it doesn't, because neither "genus" nor "family" are anywhere near defined. Without knowing anything about the palmchat (and I don't*), all I can read out of the statement that it forms a monotypic family is that it has no "close" relatives, whatever "close" might mean -- and it's highly likely to mean very different things to different people. So, my wording looks a lot more wishy-washy, but it isn't in the slightest more wishy-washy than the original.
- If given the chance, I would, as I have done here, add a statement trying to make "close" at least semiquantitative (as chemists would call it). But of course that's an addition that you could just as well make to the traditional wording. My wording just, I think, makes the need for this addition more obvious by lacking the pretend-quantification of "it's a separate family, but not a separate superfamily".
* I wrote all this before reading Darren's reply.
For example, there are people who count genera, families or orders and believe they have quantified biodiversity (for example in studies of mass extinctions or evolutionary radiations). In at least one case the very same people state in their papers that they know full well that ranks are subjective and really meaningless -- but then they do it anyway. Well, genera, families and orders are not countable. If you try to count them, you study an arbitrary cross-section through subjective opinions on splitting and lumping; actually studying biodiversity is something else.
David: That happens all the time. Since English is not your first language, I hope you will not mind if I correct you: "That happens all the damn time."
OK, my (hopefully) final comment on this thread.
I'm not so sure about that. Clearly, it doesn't need to be a paleo- vs neo-thing. But the fact remains that, for whatever reason, a randomly selected paleontologist is much more likely to be a pro-PhyloCoder than a randomly selected neontologist. I may have somewhat underestimated the proportion of neontologists among PhyloCode supporters; in particular, there are more botanists there than I was aware of. But paleontologists are represented among the key PhyloCode people way out of proportion to their actual numbers within the biological community. Consider the composition of Committee on Phylogenetic Nomenclature (2006-2007):
Tom Artois, Harold N. Bryant, David C. Cannatella, Philip D. Cantino, Julia A. Clarke, Benoit Dayrat, Kevin de Queiroz, Jacques A. Gauthier, Michel Laurin, Richard G. Olmstead, and Mieczyslaw Wolsan.
Of these eleven people, five (Bryant, Clarke, Gauthier, Laurin and Wolsan*) are paleontologists, and not just any paleontologists, but vertebrate paleontologists. That's a rather non-random distribution, is it not? And among the 'Advisory Group for earlier versions', you find yet more vertebrate paleontologists (Christopher A. Brochu, Michael S. Y. Lee, Timothy Rowe, Andre R. Wyss).
* I guess you could quibble about Wolsan, as he appears to be mainly working with Recent taxa nowadays, but he has definitely identified himself as a paleontologist in the past.
My point is that there is a strong pro-paleontology bias among those people who actually are responsible for the design of the PhyloCode. Not that that in itself is an inherently bad thing, of course; but it just might lead to the special concerns and interests of one particular field of biology being given more weight than those of many others. And because of that, I worry that this really could potentially become a paleontology vs. neontology situation at some point. Remember: there are lots of big egos involved on all sides (when aren't there?), so intellectual clashes are almost inevitable.
But yes, all this is speculation at this point. Personally, for now I am adopting a wait-and-see strategy regarding the PhyloCode. I'll keep an eye on it but I'm not going to un-learn those Linnean ranks just quite yet.
If it's a vertebrate paleontologist, yes. Otherwise the ratio might even be negative.
That has historical reasons. Hey, with AFAIK one exception, the conodont people still don't even use cladistics, and cladistic analyses of ammonites are rare, too.
The Companion Volume will have less of that bias. I know because I'm one of 13 coauthors on two chapters (Amphibia and Lissamphibia).
An important point is that the PhyloCode won't outlaw ranks altogether. They just won't have any influence on spelling, synonymy or homonymy. In other words, you can add ranks after nomenclature is done -- as opposed to the current codes, where rank assignment is part of the process of naming a taxon. Read Article 3.
Re: Cladistics was invented in 1950
I was referring to it's use in Paleontology. The way the systematics are described in a paper with the diagnostic characters, and the use of formal suffixes. idae for family, idea for superfamily, etc. It just makes for less confusion.
Could someone elaborate on the invention of cladistics in 1950? My general understanding is that cladistics as an accepted way of doing things goes back to Henning, in German in 1960, and in English in 1966. I know there were things like Wagner trees before I heard of Henning, and I vaguely recall reading of an obscure paper or two which were said to be cladistic long before Hennig.
Hennig, in German in 1950, and in English in 1966. It took that long because Hennig lived in East Berlin, and even scientific books didn't easily pass the Iron Curtain. (Plus, Hennig's style is very difficult to read in the original.)
In what respect?
-------------------
I think from the 70s onward in some subdisciplines. For dinosaurs it started in 1986 (or 1984 if conferences and dissertations count).
"Systematics are described" does not compute. What do you mean? Diagnostic characters are not part of traditional nomenclature -- the diagnosis is often called "definition", but that's actually wrong. The suffixes are part of nomenclature, but not of systematics. ~:-|
To the contrary.
Firstly, they mean that if a taxon changes rank, which of course happens whenever someone has a mood swing about which rank it "deserves", it means that its name changes. And this means that what one guy calls Iguanidae and Iguaninae is the same what another calls Pleurodonta and Iguanidae, respectively. It also means that different people use the same name for different taxa.
Secondly, they reinforce the misconception that taxa that have the same rank are somehow equivalent.
Thanks for the 1950 clarification. David Marjanovic''s comments on changes in taxon names based on increasing (or decreasing) understanding of relatioships are quite correct. It extends down to the species and subspecies level as well. It is sometimes necessary to say, to reference, what sense of the name you are using. With new ideas, we often have to have new words to express them. These new words may be just redefined old words. The author may leave it to us to puzzle out whether the old definition is used, and, if not, what is the new definition.
No, it does not. (1) Having standard endings for ranks is only informative if those ranks actually mean something, and as has been stressed by Darren and David, they do not. Darren already compared "orders" Coliiformes and Ornithischia - and that was a fairly weak example when you consider that the level of diversity covered by the average avian "order" could be quite easily contained within the average hymenopteran "subfamily", or even a moderately large angiosperm "genus". (2) I personally don't have too much of a problem with the use of different names for the same taxon - probably because, as a taxonomist myself, I work with them on a fairly regular basis. What does cause me much greater levels of grief is the use of the same name for different taxa, and the use of standard endings does not merely allow that, it directly encourages it. Consider the following brief explanatory note that I wrote when presenting a (slightly outdated now) tree for euascomycetes on Palaeos.org:
How is this less confusing?
Re: "No, it does not. (1) Having standard endings for ranks is only informative if those ranks actually mean something, and as has been stressed by Darren and David, they do not."
It makes a great deal of difference when trying to program a computer to analyze relationships. The computer does not care about ranks being equal or not but you need to compare specific characters. The problem I have is in programming for "basal" animals because it lacks systematics. It makes evolutionary analysis more difficult.
So what you're saying is that so long as yo're getting the data in, you don't care if the data actually reflects reality in any way or not?
Garbage in...
Please explain. I don't understand what ranks have to do with characters or anything else in this paragraph. ~:-|
Chris
This is not a gigo condition. A genus has a description. Most of the older papers are quite specific on the exact characters. This is easly to program. Later descriptions use similarities to other genera or similarities to family (shared) characters without being specific. So to analyze the data you need to program for every rank. What I am forced to do is make up ranks for those not properly described. Consider it a form of cladistic taxonomy. As you move up in rank you have more commonality. But all ranks are equal but more or less commonality. I'm trying to study reptilian evolution to mammals and birds so commonality is the most important factor. In the case of dinosaurs most finds are too derived to have led to birds making study of the basal genera more important.
If in fact the input is garbage it is because the description is inaccurate, like comparison to the wrong family or placed into the wrong clade, hence my complaint.