Tetrapod Zoology

In terms of its zoological diversity, Europe is the best known continent on the planet. Indeed it’s generally assumed that just about all of Europe’s macrofauna has, by now, been discovered. While that’s mostly true, it seems that at least a few species – so called ‘cryptic species’ – have been missed, mostly because they’re extremely similar to their close relatives.

i-ab74f0028ef2baa993f9df0072f3e8c2-vesper-bat-motley-wikipedia-Feb-2011.jpg

ResearchBlogging.org

Here, I want to look briefly at new discoveries made concerning the diversity of European microbats [composite above shows, clockwise from top left: Pipistrellus pipistrellus, Plecotus auritus, Pl. austriacus and Myotis nattereri. All photos from wikipedia]. Because bats are typically small and difficult to observe at close range, and because species may distinguish one another by auditory cues rather than visual ones, it now seems that many species very similar in appearance to others had been mostly ‘missed’ by traditional classifications. What typically happens in the discovery of a ‘cryptic species’ is that (1) molecular data first brings attention to its existence (often because it appears as distinctly separate on a phylogeny); (2) it’s then discovered or realised that morphological features support its distinction as well; (3) a description mentioning these (sometimes diagnostic) morphological features – sometimes published decades earlier – is found to exist, and it often includes a binomial for the ‘newly discovered’ hitherto cryptic taxon; and (4) it’s reaffirmed that neither molecular nor morphological data alone can satisfy everyone of the reality of said cryptic species, but that things are most convincing when data from both sources is combined.

i-0cb431b3847f5032d4ddd34816ce9366-Cypriot-mouse-all-over-again-Feb-2011.jpg

While most European bat species were formally named in the 1800s and before, recent work has shown that quite a few cryptic species can and do warrant recognition. In fact about 15 ‘new’ European vesper bats (‘vesper bats’ = those in the huge microbat group Vespertilionidae) have been named or resurrected from synonymy in recent decades. Readers with exceptional memories might realise that the text you’re reading here is a much-modified version of a rant I published on Tet Zoo ver 1 in 2006: this was a response to the perplexing and flat wrong statement that the Cypriot mouse Mus cypriacus [shown here] was “the first new European mammal to be discovered in 100 years”.

When it comes to the subject of recently discovered cryptic microbats, many people might immediately think of the two pipistrelle species dubbed informally the 45 and 55 kHz pipistrelles (or Common and Soprano pipistrelles), for in 1993 it was discovered that the ‘species’ Pipistrellus pipistrellus actually consisted of two distinct species, both of which differed in the echolocation frequencies of their calls. Later work showed that they differ in genetics, morphology and behaviour (Barlow et al. 1997, Davidson-Watts & Jones 2006). However, while the many differences between these two species have only recently been acknowledged, both were originally named during the 1700s and 1800s: the 45 kHz/Common pipistrelle is P. pipistellus (Schreber, 1774) while the 55 kHz/Soprano pipistrelle is P. pygmaeus Leach, 1825. Recent molecular work indicates that even the new, revised version of the Common pipistrelle consists of more than one species (Mayer et al. 2007).

Two long-eared bats no more: make that three.. four.. five.. six… seven?

Other vesper bat groups have yielded new European species comparatively recently, though as we’ll see a few of them are of controversial status. For much of the 20th century there were thought to be just two species of long-eared bat (Plecotus) in Europe: the Brown long-eared bat P. auritus (Linnaeus, 1758) and Grey long-eared bat P. austriacus (Fischer, 1829). Recent discoveries and proposals mean that there are now as many as seven, three of which occur in close proximity in the Alpine region.

i-695875addd51ada4c0fe54339f4c7826-Plecotus-macrobullaris-lip-pad-Feb-2011.jpg

Among the best understood of these is P. macrobullaris Kuzjakin, 1965, named for Swiss and Austrian long-eared bats previously regarded as intermediate between Brown long-eared bats and Grey long-eared bats (it’s now known from Croatia and elsewhere, and seems to be widespread across continental Europe). Spitzenberger et al. (2001, 2006) confirmed via molecular analysis that P. macrobullaris is worthy of species status. Like other cryptic species, it can often be diagnosed morphologically once you know what to look for: unlike other European long-eared bats, it has a small, triangular pad on its lower lip [shown here, from Tvrtković et al. 2005]. However, Ashrafi et al. (2010) noted that the pad was sometimes difficult to spot in some individuals due to its size and colour. In fact, these authors showed that many of the external morphological characters recently proposed to allow differentiation of the different long-eared bats (like fur colour and penis shape) are often unreliable: Ashrafi et al. (2010) did show, however, that comparisons of external measurements allowed the species to be distinguished extremely reliably.

A more recently named supposed species from Austria – P. microdontus Spitzenberger et al. 2002 – seems to be synonymous with P. macrobullaris (see Juste et al. 2004, Spitzenberger et al. 2006).

i-b27244e8d5f0d7c605da2ed3122d3259-Plecotus-alpinus-A-Kiefer-Feb-2011.jpg

Also recently named is the Alpine long-eared bat P. alpinus Kiefer & Veith, 2001, named for a specimen collected in France in 2001 (Kiefer & Veith 2001) but later reported from Greece, Liechtenstein, Austria, Croatia and Switzerland [P. alpinus photo shown here by A. Kiefer, from Kiefer & Veith (2001)]. The Croatian specimen was collected in 1972 and the specimen from Liechtenstein in 1961: a reminder that the actual ‘discovery’ date of a species often doesn’t match the time when it becomes technically named and/or described. P. alpinus also now seems to be synonymous with P. macrobullaris (see Juste et al. 2004, Spitzenberger et al. 2006). However, a high degree of mtDNA variation in P. macrobullaris has led to the suggestion that it might contain cryptic species (Mayer et al. 2007), and it’s possible that ongoing work will show that these cryptic taxa – if they warrant recognition – correspond to P. microdontus or P. alpinus.

In 2002, another new one – the Sardinian long-eared bat P. sardus Mucedda et al., 2002 [shown below; photo by Mauro Mucedda, from wikipedia] – was described from Sardinia. Local bat experts had long thought that these Sardinian bats were morphologically distinctive compared to other European long-eared bats, but their suspicions were only confirmed after molecular results were obtained. P. sardus seems to be more closely related to the Brown long-eared bat than to species like P. macrobullaris, and it’s said that it can be identified in the hand thanks to the proportional size of its tragus, foot and thumb (among other features).

i-751a514bdcfe16fca9b5e16fa9daa56d-Plecotus-sardus-Mauro-Mucedda-wikipedia-Feb-2011.jpg

Several additional new long-eared bat taxa have also been named within the last few decades: originally named as subspecies, new data has caused some of them to be elevated to species level. Within P. auritus, the subspecies P. a. hispanicus (later reidentified as a subspecies of P. austriacus) was named in 1957, P. a. kolombatovici in 1980, and P. a. begognae in 1994. Genetic studies have since shown that both P. a. kolombatovici from Croatia and P. a. begognae from the Iberian Peninsula are distinct enough to be regarded as full species (Mayer & von Helverson 2001, Spitzenberger et al. 2001, Mayer et al. 2007), though the latter taxon is not widely accepted at the moment.

Another form first named as a subspecies of P. auritus, P. a. teneriffae Barret-Hamilton, 1907 was elevated back to species status in 1985 (see also Spitzenberger et al. 2006). Though originally named as a Canary Islands endemic, bats referred to this species were later reported from north Africa and the Balkans. However, it now seems that they belong either to P. kolombatovici or P. austriacus.

The final list of currently recognised species is therefore P. auritus, P. austriacus, P. kolombatovici, P. macrobullaris, the Sardinian endemic P. sardus, and the Canary Islands endemic P. teneriffae. Incidentally, north African bats until recently included in P. kolombatovici are now regarded as distinct enough for their own species: P. gaisleri Benda et al., 2004.

New mouse-eared bats aplenty?

Another new vesper bat, this time a mouse-eared bat (= Myotis), was named during the 1970s but now seems unlikely to be a valid species. It’s the Nathaline bat Myotis nathalinae Tupinier, 1977, described for two specimens from Ciudad Real in Spain. However, it’s highly similar genetically and morphologically to Daubenton’s bat M. daubentonii (Tupinier 1977). Indeed Bogdanowicz (1990) found that the skull morphology of M. nathalinae fell within the range of variation exhibited by M. daubentonii populations, and therefore argued against the idea that it should be regarded as a valid species. Genetic samples of M. nathalinae have also fallen within the range of variation exhibited by M. daubentonii (Mayer & von Helverson 2001). Other studies have produced the same result, so bat workers generally regard M. nathalinae as a subspecies of M. daubentonii.

Alcathoe bats everywhere!

i-e64177c0dc4a3272198523a23391aef5-Alcathoe's-bat-BBC-News-Feb-2011.jpg

A second new mouse-eared bat, M. alcathoe von Helverson et al., 2001, is morphologically and genetically distinct, and noteworthy in being Europe’s smallest mouse-eared bat, and the one with the most high-pitched echolocation calls. It’s now generally known as the Alcathoe bat (the name ‘Alcathoe’s bat’ is used a lot, but is – so I’m told – incorrect). First reported from Greece, it was later reported from Slovakia, Germany, Poland, Albania, Turkey and, most recently, from two widely separate locations in the UK (Jan et al. 2010). It seems to be a widely distributed but rare species limited to moist deciduous forests with old trees and running water. In reviewing its distribution as of 2007, Niermann et al. (2007) suggested that it might be discovered even further north than Germany: a prediction borne out by its discovery in Yorkshire, England [adjacent M. alcathoe photo © Cyril Schönbächler].

Judging from the results of Mayer et al. (2007), some additional European mouse-eared bat species possibly await recognition. Bats from Austria and Italy – identified for morphological reasons as Natterer’s bat M. nattereri – were significantly different in mtDNA from other Natterer’s bat samples, while Bulgarian bats identified as individuals of the Whiskered bat M. mystacinus carried a distinct mtDNA haplotype relative to ‘true’ M. mystacinus (Mayer et al. 2007). Bulgaria is already home to two mouse-eared bat populations that have sometimes been confused but do seem distinct, namely M. mystacinus bulgaricus and M. aurascens. The latter – sometimes called the Steppe whiskered bat – was named in 1935 as a subspecies of the Whiskered bat. It seems to be the commonest mouse-eared bat on the Balkan Peninsula, being known from Albania, Bulgaria, Greece, Serbia, Kosovo, Macedonia and Montenegro (Benda 2004).

Even this isn’t the end of it, as there are additional bat populations on the Canary Islands and on some of the Greek islands that have proved substantially different in genetics from the other populations conventionally included with them in the same species. In the cladogram below – from Mayer et al. (2007) – the species first recognised thanks to molecular techniques have their names placed within white rectangles, while the new cryptic species discovered in this one study are signified by the black rectangles.

i-ca640a3f62d2ba791efa1e1c265bf4d8-Mayer-et-al-2007-vesper-bat-phylogeny-Feb-2011.jpg

So far, recent studies on European vesper bats have resulted in the proposed existence of more than 15 hitherto overlooked bat species in the region, and continuing studies indicate that even more await recognition. Some of the bats discussed here have also been discovered using more ‘traditional’ means: that is, by the discovery of individuals that look different from their close relatives, and are then judged to warrant specific status. We might regard it as unbelievable and amazing that Europe (of all places) continues to yield such species. On the other hand, the discovery of new tetrapod species isn’t a rare thing (something I’ve tried to emphasis here on Tet Zoo), and maybe we shouldn’t be surprised at all.

Finally, I know that a lot of people are sceptical when it comes to the recognition of cryptic species and sometimes insist that there’s little point in recognising them as distinct, nameable entities. However, this ignores the important facts that (1) while some species might look very much alike, it’s not looks alone that matter, and (2) those cryptic species are frequently relatively old and relatively distinct (genetically at least) relative to their close relatives, frequently being as old and as distinct – if not more distinct – than many traditionally recognised species.

For previous Tet Zoo articles on microbats see…

And for more on recently discovered mammals, see…

Refs – -

Ashrafi, S., Bontadina, F., Kiefer, A., Pavlinic, I. & Arlettaz, R. 2010. Multiple morphological characters needed for field identification of cryptic long-eared bat species around the Swiss Alps. Journal of Zoology doi:10.1111/j.1469-7998.2010.00697.x

Barlow, K., Jones, G., & Barratt, E. (1997). Can skull morphology be used to predict ecological relationships between bat species? A test using two cryptic species of pipistrelle Proceedings of the Royal Society B: Biological Sciences, 264 (1388), 1695-1700 DOI: 10.1098/rspb.1997.0235

Benda, P. 2004. First record of Myotis aurascens and second record of Myotis brandtii in Montenegro. Lynx 35, 13-18.

Bogdanowicz, W. 1990. Geographic variation and taxonomy of Daubenton’s bat, Myotis daubentoni, in Europe. Journal of Mammalogy 71, 205-218.

Davidson-Watts, I. & Jones, G. 2005. Differences in foraging behaviour between Pipistrellus pipistrellus (Schreber, 1774) and Pipistrellus pygmaeus (Leach, 1825). Journal of Zoology 268, 55-62.

Jan, C. M. I., Frith, K., Glover, A. M., Butlin, R. K., Scott, C. D., Greenaway, F., Ruedi, M., Frantz, A. C., Dawson, D. A. & Altringham, J. D. 2010. Myotis alcathoe confirmed in the UK from mitochondrial and microsatellite DNA. Acta Chiropterologica 12, 471-483.

Juste, J., Ibáñez, C., Muñoz, J., Trujillo, D., Benda, P., Karatş, A. & Ruedi, M. 2004. Mitochondrial phylogeography of the long-eared bats (Plecotus) in the Mediterranean Palaearctic and Atlantic Islands. Molecular Phylogenetics and Evolution 31, 1114-1126.

Kiefer, A. & Veith, M. 2001. A new species of long-eared bat from Europe (Chiroptera: Vespertilionidae). Myotis 39, 5-16.

Mayer, F. & von Helversen, O. 2001. Cryptic diversity in European bats. Proceedings of the Royal Society of London B 268, 1825-1832.

- ., Dietz, C. & Kiefer, A. 2007. Molecular species identification boosts bat diversity. Frontiers in Zoology 4, 1-4.

Niermann, I., Biedermann, M., Bogdanowicz, W., Brinkmann, R., Le Bris, Y., Ciechanowski, M., Dietz, C., Dietz, I., Estók, P., Helversen, O. V., Le Houédec, A., Paksuz, S., Petrov, B. P., Özkan, B., Piksa, K., Rachwald, A., Roué, S. Y., Sachanowicz, K., Schorcht, W., Tereba, A. & Mayer, F. 2007. Biogeography of the recently described Myotis alcathoe von Helversen and Heller, 2001. Acta Chiropterologica 9, 361-378.

Spitzenberger, F., Piálek, J. & Haring, E. 2001. Systematics of the genus Plecotus (Mammalia, Vespertilionidae) in Austria based on morphometric and molecular investigations. Folia Zoologica 50, 161-172.

- ., Strelkov, P. P., Winkler, H. & Haring, E. 2006. A preliminary revision of the genus Plecotus (Chiroptera, Vespertilionidae) based on genetic and morphological results. Zoologica Scripta 35, 187-230.

Tupinier, Y. 1977. Description d’une Chauve-souris nouvelle: Myotis nathalinae nov. sp. (Chiroptera, Vespertilionidae). Mammalia 41, 327-340.

Tvrtković, N., Pavlinić, I. & Haring, E. 2005. Four species of long-eared bats (Plecotus, Geoffroy, 1818; Mammalia, Vespertilionidae) in Croatia: field identification and distribution. Folia Zoologica 54, 75-88.

Comments

  1. #1 Andreas Johansson
    February 5, 2011

    I find it a tad difficult to think of animals from Cyprus or the Canaries as “European”.

    Anyway, it’s neat how molecular studies prompt the (re)discovery of previously overlooked morphological distinctions.

  2. #2 Lassi Hippeläinen
    February 5, 2011

    Cyprus is definitely in Asia in terms of geology and biology. Politics is different. But if you accept political arguments, you must extend Europe to Tahiti.

  3. #3 David Marjanović
    February 6, 2011

    Mayer et al. (2007)

    A neighbor-joining tree in 2007? That’s sad.

    Cyprus is definitely in Asia in terms of geology

    What is Asia in terms of geology? I mean, you’re not talking about Angara, but what do you mean?

  4. #4 Tommy Tyrberg
    February 6, 2011

    Cyprus is a piece of uplifted Tethys seabottom, on the Eurasian side of the plate margin. However on this basis one could just as easily argue that Arabia and India are not part of Asia or that Maghreb is part of Europe.

  5. #5 Vladimir Dinets
    February 6, 2011

    I think the tiny number of comments is a nice illustration of how blissfully under-appreciated small mammals are compared to birds.
    A description of a cryptic species in birds follows a completely different sequence of steps:
    (1) Someone notices a small dialect difference between vocalizations of two geographic populations, or slightly non-random mating in 1000 km-wide intergradation zone, or finds that genetic distance between two subspecies is the same as between two unrelated forms that someone somewhere has considered full species.
    (2) Birding mailing lists fill with excited posts titled “Rejoice! New rumors of possible splits!”
    (3) The “new species” is included into some twitching software package or some obscure regional checklist.
    (4) Some birding committee or association votes to recognize the new species, knowing all too well that any dissenter would be burnt at stake by an angry mob, and that refusal to split might cause unhappy twitchers to take their membership fees elsewhere.
    (5) Immediately, five other populations get split in the same way, using this first split as justification.

  6. #6 Andreas Johansson
    February 6, 2011

    Vladimir Dinets wrote:

    blissfully under-appreciated

    Now there’s an expression I need to use more often.

  7. #7 Jerzy
    February 6, 2011

    On the other hand, small mammals are split like this:
    1) Somebody finds a strange bat, decides he is sober enough today so it is different than in textbooks, so splits it.
    2) Nobody else cares for a next decade/century.

    BTW, what are criteria of “species” in bats, which these things met?
    And did anybody ever, ever made a study of hybridization or lack thereof between two similar bat species?

  8. #9 Allen Hazen
    February 6, 2011

    Compared to birds and bats, small terrestrial mammals don’t travel very far. So (I would guess) it’s comparatively easy for populations of them to become reproductively isolated, so (one would think) there would be LOTS of cryptic species of them. (Sort of like why more language families are found in the Caucasus or Papua New Guinea where getting from one valley to the next is hard than in an equivalent area of, say, the central Eurasian steppe.) I think Darren has blogged on this, perhaps back in Version 1. Certainly I recall reading about places (in Brazil?) where you get different species of mouse on the two sides of a big river.

    But now we see that cryptic speciation is rampant in VOLANT mammals. My (amateur’s) suspicion is that this suggests that cryptic speciation is even MORE rampant in small terrestrial mammals. Has anyone tried to estimate what the true number of species of, say, moles is? I wouldn’t be surprised if what the nature guides list as one species of mole ranging over half a continent, if studied molecularly and then morphologically, turned out to be a dozen!

    (Have moles been studied much? I know there is a literature on shrews, of which there seem to be a lot more species than the nature guides of my childhood recognized.)

  9. #10 Christopher Taylor
    February 7, 2011

    And did anybody ever, ever made a study of hybridization or lack thereof between two similar bat species?

    That is (theoretically) what population genetic studies are supposed to be testing: whether or not introgression is occurring between populations. I say ‘theoretically’ because there have been some supposed population genetic studies (not necessarily the bat ones, just speaking in general) that could be argued to not have the necessary coverage to properly test this.

    Unless you specifically meant direct breeding tests, in which case it should be pointed out that such studies (whether or not they’ve been done in the past) are all too often impractical, inadvisable and inconclusive. Impractical and inadvisable because many animals are exceedingly difficult to maintain in captivity in conditions conducive to breeding (which would be both practically and ethically required of researchers), and because the number of breeding tests that would be required is unfeasible. Inconclusive because, even if you do get results, there is always grounds for questioning whether behaviour in captivity is a proper reflection of behaviour in the wild. If your bat-breeding results are negative, does that indicate that the individuals belong to different species – or do they indicate that that particular female wouldn’t mate with that particular male if he was the last pipistrelle on earth? If the results are positive, does that indicate that the individuals belong to the same species – or are they the result of lack of choice leading to the acceptance of a normally unacceptable sexual partner?

  10. #11 Dartian
    February 7, 2011

    Darren:

    Even this isn’t the end of it

    Indeed it isn’t. You missed at least one ‘new’ mouse-eared bat species from Europe: Myotis punicus. Castella et al. (2000) found no evidence of gene flow between greater mouse-eared bats Myotis myotis from southern Spain and their supposed conspecifics on the other side of the Gibraltar Strait in Morocco. Their genetic data suggest that the Moroccan, and also the Corsican and the Sardinian, mouse-eared bats are specifically distinct from those on the Iberian Peninsula and the European mainland. In a later study, Evin et al. (2008) found morphological support for that view; their data also suggest that within Myotis punicus, the populations on Corsica and Sardinia might warrant further species-level splitting.

    References:

    Castella, V., Ruedi, M., Excoffier, L., Ibáñez, C., Arlettaz, R. & Hausser, J. 2000. Is the Gibraltar Strait a barrier to gene flow for the bat Myotis myotis (Chiroptera: Vespertilionidae)? Molecular Ecology 9, 1761-1772.

    Evin, A., Baylac, M., Ruedi, M., Mucedda, M. & Pons, J.-M. 2008. Taxonomy, skull diversity and evolution in a species complex of Myotis (Chiroptera: Vespertilionidae): a geometric morphometric appraisal. Biological Journal of the Linnean Society 95, 529–538.

    Vladimir:

    Some birding committee or association votes to recognize the new species, knowing all too well that any dissenter would be burnt at stake by an angry mob, and that refusal to split might cause unhappy twitchers to take their membership fees elsewhere.

    The fabulously wealthy, all-powerful Birdwatcher Lobby decides these things by the threat of financial blackmail? Really now?

    How about supporting your paranoid-sounding scenario by giving a few real-world examples (properly documented and referenced, of course) of bird species that have been erected under such circumstances as you described?

  11. #12 Darren Naish
    February 7, 2011

    Thanks for great comments. On hybridisation or lack of (comment 7) within these bats, it has indeed been reported in some European vesper bats (Berthier et al. 2006). Incidentally, some of the taxa discussed here (like Plecotus macrobullaris) were initially reported – erroneously – as hybrids. Otherwise, I don’t think hybridisation has been reported in the species discussed above (of course, while hybridisation is easy to spot in big animals, it’s difficult to (1) observe and (2) confirm in micro-mammals). And, as noted in the article, two and possibly three of the long-eared bats discussed above occur in close proximity, and two definitely share roosts. Common and Soprano pipistrelles also share roosts and seem not to hybridise, with some studies indicating that the two are reliably able to distinguish one another based on odour.

    More importantly, I think it’s misleading to imply that hybridisation means anything anyway – as is increasingly recognised (it’s been mentioned here quite a few times), the populations that we regard as ‘species’ frequently do hybridise with others; ergo, hybridisation does not necessarily provide any reliable indication of whether a species is a ‘good’ one or not. The units here regarded as ‘species’ have not been proposed on the basis of a biological species concept: as should be more widely known, the BSC has never been much employed by those who actually name species (despite what many [mostly ecology- or behaviour-themed] popular books say). Rather, these species are ‘diagnostic species’, recognised as distinct on the basis of diagnostic characters.

    And I had indeed missed the memo on Sardinian/Corsican/Moroccan Myotis – thanks for the refs, Dartian.

    Ref – -

    Berthier, P., Excoffier, L. & Ruedi, M. 2006. Recurrent replacement of mtDNA and cryptic hybridization between two sibling bat species Myotis myotis and Myotis blythii. Proceedings of the Royal Society B 273, 3101–3109.

  12. #13 Don Cox
    February 7, 2011

    The differences in sound frequencies suggest a preference for different sizes of insect prey.

    If we had a useful sense of smell, we might well find that these closely related species are easily distinguished by smell. If so, that would discourage cross-breeding.

  13. #14 Pedro Cardia
    February 7, 2011

    Hi,

    Allen Hazen’s comment (#9) just brought to my memory the recent “explosion” in the number of recognized species within the genus Podarcis (Reptilia) in the Iberian Peninsula. This is still ongoing work, but a lot of it stemmed from molecular and morphological work done in the last 10 years.
    Reptiles are even less mobile than micro-mammals…

  14. #15 Vladimir Dinets
    February 7, 2011

    Dartian: that’s easy. Look at the recent ABA decision to split snowy and Kentish plovers, look at all those splits of Neotropic owls based exclusively on dialect differences, look at recent splits of West Indies warblers based exclusively on the fact that their genetic differences are larger than between some North American ones which shouldn’t have been recognized as full species in the first place. On the other hand, a paper showing that Northern Parula is a subspecies of Tropical was ignored so completely, it’s not even mentioned in the last HBW volume.

  15. #16 Dartian
    February 7, 2011

    Look at the recent ABA decision to split snowy and Kentish plovers, look at all those splits of Neotropic owls based exclusively on dialect differences, look at recent splits of West Indies warblers based exclusively on the fact that their genetic differences are larger than between some North American ones which shouldn’t have been recognized as full species in the first place. On the other hand, a paper showing that Northern Parula is a subspecies of Tropical was ignored so completely, it’s not even mentioned in the last HBW volume.

    But which of those instances of splitting resulted from twitchers threatening to resign en masse and ‘take their membership fees elsewhere’? You specifically stated that as a reason.

  16. #17 David Marjanović
    February 7, 2011

    Some birding committee or association votes to recognize the new species

    what is this I don’t even

    “Recognize”???

    Science by democracy???

    If we had a useful sense of smell, we might well find that these closely related species are easily distinguished by smell.

    I bet we actually can do it. (Except presumably for the smokers among us.)

  17. #18 Darren Naish
    February 7, 2011

    There are a few groups of tetrapods where a distinctive aroma has indeed been mentioned (often repeatedly) as a diagnostic feature, and I do mean by the human describers of such animals. Hawaiian honeycreepers (drepanidines), for example, have a unique ‘honeycreeper odour’, and the absence of such odour from some controversial taxa has even been used to bolster the idea that they should be removed from the group. I’m pretty sure I remember reading about a group of frogs where people were able to distinguish genera (at least) by odour. Note that I’m not referring to obvious odoriferous taxa like skunks, pungent colubroid snakes and so on.

  18. #19 Darren Naish
    February 7, 2011

    … and, to clarify, I wasn’t talking about distinguishing species based on odour, though this might be possible in cases.

  19. #20 Vladimir Dinets
    February 7, 2011

    Dartian: How exactly do you expect me to provide documentation of this? We both know that lumping papers get ignored and splitting papers are eagerly accepted by the community which I can no longer call scientific. Why do you think it happens? Is it because most ornithologists are twitchers themselves nowadays, or because of the pressure of knowing what the public wants from you? Does it really matter? The number of bogus species keeps growing, that’s all that matters.
    Darren: Identification by smell used to be very popular in botany in the old days. I think similar smell was among the original reasons for lumping skunk-badgers with skunks (but convergently similar zorillas smell different). I wonder if dogs could be trained to distinguish between species, just like they can be trained to sniff out cancers.

  20. #21 Sven DiMilo
    February 7, 2011

    These probably qualify as “obviously odiferous” but I used to be able to tell a Sternotherus from a Kinosternon by…aroma.

  21. #22 Dartian
    February 7, 2011

    How exactly do you expect me to provide documentation of this?

    If you can’t back up your claims, then you shouldn’t be making those claims in the first place. Allegations of academic misconduct are serious matters – not something that should be thrown around like confetti.

  22. #23 Jerzy
    February 7, 2011

    I would agree with Vladimir that birds are now oversplit, but wrongly understood conservation politicis is more important than twitching.

    About races etc.
    I remember that somebody studied chromosomal races of shrews (note that shrews with different chromosomes can hybridize, despite what you learned on genetics class). The result was 8 shrew forms in a country size of Poland.
    I also remember that naked mole rats populations close by are as genetically different as subspecies of other rodents. On the other hand, who would care about 100 species of naked mole rat?

    About smell: there is a legend that one scientist was able to recognize scats (sh*t) of House Marten and Pine Marten by smell.
    Some fungi are also recognized by smell.
    I guess human nose could be trained to recognize many similar mammal species by smell. One question – how to develop comparison chart? And can smells be amplified to human nose?

  23. #24 Darren Naish
    February 7, 2011

    It now seems that polymorphism in chromosome number is the norm in wild boar – we discussed this on Tet Zoo recently actually (see the comments under this article). It’s also true of Moose, domestic cattle and Okapi and might be true of quite a few other artiodactyls. Among mammals, some rodents (Ctenomys and various rats, mice and voles) are also chromosomally polymorphic.

  24. #25 Adam F
    February 7, 2011

    I had a phylogenetics prof who would have said that birds are highly lumped. Of course, he is a fish phylogenist and thinks any distinguishable population should be considered a species of its own.

  25. #26 Vladimir Dinets
    February 7, 2011

    Dartian: I provided you with examples of biased approach. Precise motifs are not that important. You want more cases of ignored lumps and bogus splits? Pretty much anywhere you look: Parulids, juncos, rosy-finches, parakeets, almost every Australasian bird that happens to live on more than one island, the list goes on forever.
    I wouldn’t call it “Academic misconduct”, because AOU and ABA and organizations like that are not scientific. Science is not done by vote. The problem is that their decisions become the new dogma, and dogma in science is never a good thing.

  26. #27 Dartian
    February 8, 2011

    Vladimir:

    I provided you with examples of biased approach

    You did no such thing. You said ‘lumping papers get ignored and splitting papers are eagerly accepted’. That’s just an assertion. If there is a systematic bias there somewhere, merely listing cases of splitting isn’t enough to demonstrate that. Show your statistics. For example:

    -In the last XX years, how many studies have been published that suggest species-level taxonomic changes in birds?
    -In how many cases have these studies recommended splitting, and in how many cases have they recommended lumping?
    -How many of these suggested changes have actually been accepted (and ‘made official’) by ornithological societies in various countries?
    -How many of these suggested changes have been rejected? For which (if any) stated reasons? Do instances where the taxonomic status quo is retained outnumber instances where splits and/or lumps are accepted?
    -Are there notable differences between different countries and organisations? Do, for example, financially poorer ornithological societies accept splits more readily than wealthier organisations? (They should, if your ‘we-won’t-pay-our-membership-fees-if-you-don’t-split-these-taxa’ claim is true.)

    And so on, and so forth. Anyone can make assertions. But in science, that’s not enough; you need to crunch the numbers too. If/when the data show a significant bias favouring species-level splitting to lumping and/or retaining the taxonomic status quo, then – and only then – should you start speculating on what causes this bias.

    Precise motifs are not that important.

    It was you who dragged motives [sic; in English, 'motif' and 'motive' are two different things] into this discussion by claiming that bird species are split for financial and PR reasons. If motives aren’t relevant, then they shouldn’t have been brought up either.

  27. #28 Darren Naish
    February 8, 2011

    For those who care… I learnt last night that ‘Alcathoe’ is pronounced “al-kaa-toe-ee”, apparently.

  28. #29 Vladimir Dinets
    February 8, 2011

    Dartian,
    You expect me to write a full-time scientific paper and publish it as a blog comment? Sorry, I really don’t have time now – my thesis defense is in six weeks.
    But if you want statistics, here is a little bit of data to play with. There’s been a number of papers proposing lumping North American species in the last 5 years. Rosy-finches, parulas, a few others. As far as I remember, the only one that was accepted by AOU etc. was the lumping of green-winged teal – but that one had never been really recognized as a valid split prior to that. Most weren’t even rejected – they were ignored. So the acceptance rate for lumps is zero.
    For splitting papers it’s definitely not zero. I would have to go through a lot of papers to get 100% coverage, but I strongly suspect the p-value would be pretty low. You can do it yourself if you have time.
    And sorry, English is not my first language, so I do stupid typos sometimes. I don’t think it prevented anyone from understanding the idea this time.

  29. #30 Dartian
    February 8, 2011

    English is not my first language, so I do stupid typos sometimes.

    Sorry about pointing it out, but my SIWOTI syndrome compelled me to do it. And if it’s any consolation, some of my Tet Zoo typos have been much more stupid than yours.

  30. #31 John Harshman
    February 8, 2011

    It isn’t enough to show that there are more splitting papers than lumping papers, or that more splitting proposals than lumping proposals are accepted by the AOU checklist committee. You have to show that they were done wrongly. Perhaps birds were previously over-lumped, and this is now being corrected. Certainly “Northern oriole” is a good example of that. There was a late 19th Century fashion for splitting, and a mid-20th Century fashion for lumping in reaction. Which went too far.

    I think a much more important contribution to splitting than any sort of twitcher conspiracy is twofold: first, the great amount of molecular data now available that can help detect cryptic species, and second, the current popularity among ornithologists of a phylogenetic species concept.

    So what changes don’t you like, and why?

  31. #32 Vladimir Dinets
    February 8, 2011

    John: lumping proposals are not accepted, period. They are just ignored. I gave two examples: parulas and rosy-finches. There was a bunch of others over the years. A list of poorly justified splits is very long, and many of them don’t have any molecular evidence. Any bird that happens to have a slightly differing population in NE Brazil or the central Andes of Peru is automatically split nowadays.
    The question is not whether there is bias or not. Among people working in low-level taxonomy, there is always a strong bias towards splitting, for a number of reasons. This bias has to be countered by people looking at higher levels – doing large-scale faunistic or taxonomic reviews. This mechanism just no longer works in ornithology, and the result is runaway splitting. There are now 130 families in Passerines – a clade so uniform that even splitting it into two or three families might not be justified. The most basal extant group – the New Zealand wrens – looks and lives exactly like some of the most advanced ones. This madness is nicely chronicled at http://creagrus.home.montereybay.com/list.html

  32. #33 farandfew
    February 8, 2011

    I don’t know about regional checklists, or twitching software packages (what are they?) or about how things work in the US in general. But, with field guides and with the species lists accepted by conservation organisations, I’d be extremely surprised if there were not a bias in favour of accepting splits and rejecting lumps.
    In fact this is rational behaviour for birdwatchers and conservationists even though it’s quite wrong for taxonomists or scientists using taxonomies. It’s rational because it follows the precautionary principle. If a birdwatcher knows how to identify members of a population which may or may not be a valid species then if, in the end, it turns out not to be valid; he/she has lost nothing much (though has perhaps lost some time and money if he/she actually went to look for the thing). If, on the other hand, it turns out to be valid, he/or she has gained a new species for his/her list. The same is true, in a much more serious way, of conservationists.
    I say ‘it turns out to be’ a valid species because people who write field guides or species lists for NGOs have to work on the principle of scientific consensus rather than on evaluating the evidence themselves. In the latter case they don’t have time to read the papers. In the former, they will indeed get complaints from reviewers that certain possible new species were left out.

    …not that the reverse kind of complaint is impossible, or that birders in general react with joy to news of splits. Maybe they do in America but most birders I know seem to feel it makes the whole thing ridiculous (not to mention difficult). Even so, I think they’d prefer to have the split species in there than not.

  33. #34 Darren Naish
    February 8, 2011

    Remember that ‘family’ does not mean anything special anymore: all those new (or resurrected) passerine ‘families’ are just named clades, or sometimes named lineages containing a single species. It’s a historical vestige of the Linnaean system that many people (even those perfectly au fait with cladistics) still want to create ‘higher groups’ for taxa that stand alone relative to others.

    As for the whole splitting debate on species, it’s all too easy to forget that the term ‘species’ can mean so many different things. I note that those who object to splitting proposals often have a vague, amorphous and very ‘inclusive’ concept of what a species should be; a concept that is quite different from that understood by the specialists who work on the groups that cause the arguments. I remember reading a good commentary by Nigel Cleere on how Old World nightjars had been over-lumped (it was online, and unfortunately I can’t find it anymore). Time and time again, he was able to show that the lumping proposals (1) had been absolutely lazy, being the result of quick conclusions needed for major synthetic works, (2) had ignored lots of preceding careful, detailed work where the relevant diagnostic characters were explained and illustrated well, and (3) were absolutely discordant with new data from vocalisations, moults and DNA, some of which showed that the populations previously lumped as conspecifics were not close relatives at all.

    And I never really understand the objections to the raising of distinctive ‘subspecies’ to species status given the long, previous, unquestioned acceptance of such things as umpteen Dendroica and Empidonax species.

  34. #35 Vladimir Dinets
    February 8, 2011

    Darren,
    A family might not mean much nowadays, but it is misleading to have different standards applied to different groups. People who try to estimate biodiversity, or the evolutionary success or a certain group, assume that five families of birds are more or less equal to five families of lizards, even though it reality taxonomic levels in birds are by now mostly one level up compared with everybody else. The level of morphological and life history diversity among extant Aves is more or less the same as within large mammalian or reptilian orders, if not less. And it goes all the way down to genera. It is probably unavoidable that vertebrates will always be oversplit compared to invertebrates, but at least we should try to be consistent within vertebrates, don’t you think?
    And yes, some Dendroica and Vermivora should be lumped, not to mention that they should be merged together and with a bunch of other genera, but what are the chances of anyone devoting their time to writing a paper that is certain to be ignored?

  35. #36 Darren Naish
    February 8, 2011

    Wait a minute… the standards applied to different tetrapod groups are _totally_ different! ‘Families’ and ‘Orders’ and so on are completely inconsistent across the major tetrapod groups – I think they’re downright unhelpful, as people assume that these ranked groups are somehow equivalent when they definitely aren’t. But I’ll stop there, as we might spin off at a tangent (this isn’t the same as the whole cryptic species/splitting debate).

  36. #37 Jerzy
    February 8, 2011

    In Western Palearctic birds, pretty much all changes in the last 20 years were splitting subspecies into species. About 20-40 species so far.

    One shameful extreme (and only lumping I can think of) was spliting of Lesser Black-backed Gull Larus fuscus into fuscus, graellsii and heuglini, which was reversed only after “species” turned to be unidentifable except when ringed as chicks within known breeding range of a given form.

  37. #38 Knightly
    February 8, 2011

    Excellent. We require more Vespene Bats.

    Also pylons.

  38. #39 John Harshman
    February 8, 2011

    Let’s face it: a family never meant anything, not even in the good old days. There is no definition by which a family can be distinguished from a genus or an order. It was just some taxonomists idea of a nice group. Many of them are paraphyletic or worse. If people assume equivalence of families, or any other ranks, they’re just being silly. Blame them, not the taxonomy.

    All this is of course a completely different rant from your rant on species splitting. Species at least have definitional criteria (too many criteria, but nothing is perfect).

    And I’ve seen the data on which rosy finches were supposed to be lumped; they don’t support lumping.

  39. #40 Jerzy
    February 8, 2011

    BTW, African nightjars might well be undersplit. This is another failing of bird taxonomy – they ignore really interesting cases in difficult groups and regions (S American suboscine passerines are another example) but concentrate on high-profile species in well twitched areas – like Europe and Canary islands.

    Somebody ranted aout the same in reptile taxonomy once.

  40. #41 Jerzy
    February 8, 2011

    BTW – I like Vladimir’s radical idea to reduce number of families in passerines. Maybe not to 2-3, but 10-15 would do.

    Another clincher that monophyletic clades are useless in zoology. To have few clades which are well-defined biologically and ecologically and really worth using, one must cope with lots of near-identical and uninformative clades.

  41. #42 John Harshman
    February 8, 2011

    Jerzy, Purely from curiosity, which 10-15 passerine families would you keep? Perhaps a nice, ecologically uniform family like “9-primaried oscines”?

  42. #43 Vladimir Dinets
    February 8, 2011

    Darren: Yes! Not all families are equal. But why make it even worse than it already is?

    John: It is not a rant. It is a heroic attempt to draw people’s attention to some inconvenient truth that the evil alien-controlled totalitarian conspiracy… sorry, wrong browser window… some inconvenient truth that has to be recognized if we want to avoid a total loss of sense in avian systematics. As for rosy-finches, I don’t have it at hand, but you can look it up in rosy-finch accounts in the last HBW volume (I think Lynx Editions should start a free radio station to bring a word of truth to American birders, oppressed and brainwashed by AOU… just kidding, HBW isn’t much better).

    Jerzy: I tried all possible ways of splitting Passeriformies into a reasonable number of families, but anything other than 1 gives you families that don’t look natural, and lots of intermediate forms that don’t fit. You can’t even break it into suboscines and oscines, as suboscines are paraphyletic and there’s nowhere to put lyrebirds and the like. If it’s more than 2 families, than the first ones you have to split are New Zealand wrens… at which point the exercise becomes ridiculous anyway.

  43. #44 Darren Naish
    February 9, 2011

    We’re talking about a hugely diverse group of over 5000 species. I don’t wish to sound rude, but the idea that you can somehow carve them up into just 15 or less ‘families’ is ridiculous and not to be taken seriously. Anyway, if you do want to talk about larger assemblages, remember that ‘families’ are not the only clades we recognise: if you want to break up passerines in the ‘lumpiest’ way possible, how about (1) acanthisittids, (2) eurylaimides, (3) tyrannides, (4) menuroids, (4) meliphagoids, (5) ‘core corvoids’, (6) petroicids, (7) passeroids, (8) stenostirids, (9) regulids, (10) ‘picathartoids’, (11) bombycillids, (12) parids, (13) muscicapoids, (14) certhioids and (15) sylvioids. This still leaves out quite a few taxa that are of uncertain placement.

  44. #45 David Marjanović
    February 9, 2011

    my thesis defense is in six weeks

    Congratulations in advance! :-) What is your thesis about?

    There are now 130 families in Passerines – a clade so uniform that even splitting it into two or three families might not be justified.

    It has been said a few times in this thread, but it cannot be said often enough:

    “Justified” does not apply to “families”.

    “Justified” does not even apply to genera.

    “Justified” applies even to species only if you have already decided which species concept to use.

    A family might not mean much nowadays, but it is misleading to have different standards applied to different groups.

    There aren’t different standards for different groups.

    There are no standards for any groups.

    People who try to estimate biodiversity, or the evolutionary success or a certain group, assume that five families of birds are more or less equal to five families of lizards

    Biodiversity must not be estimated by counting families or genera, because families and genera are not countable.

    Maybe it would make sense to estimate biodiversity by counting species after having applied one and the same species concept throughout the group in question. But this has never been done.

    The best existing way to estimate biodiversity is the Phylogenetic Diversity Index. It consists basically of adding branch lengths on a phylogeny. Yes, you need a phylogenetic tree first; nothing in biology makes sense except in the light of evolution.

    it reality taxonomic levels in birds are by now mostly one level up compared with everybody else

    There is no such thing as a level.

    Except, arguably, for the species level if and only if you apply one single species concept to everything — which nobody has ever done.

    The level of morphological and life history diversity among extant Aves is more or less the same as within large mammalian or reptilian orders, if not less.

    How did you measure this?

    You didn’t. You didn’t even try, because it’s not possible.

    Another clincher that monophyletic clades are useless in zoology. To have few clades which are well-defined biologically and ecologically and really worth using, one must cope with lots of near-identical and uninformative clades.

    Please explain.

    And please stop saying “monophyletic clade” — clades are by definition monophyletic, “clade” and “monophylum” are synonyms, “clade” is defined as “an ancestor and all its descendants”.

  45. #46 Christopher Taylor
    February 9, 2011

    To have few clades which are well-defined biologically and ecologically and really worth using, one must cope with lots of near-identical and uninformative clades.

    Not necessarily. This attitude is actually a hang-over from ranked systems, where recognition of a taxon at a particular rank demands that there also be a corresponding taxon at that rank for its sister group(s). An unranked system doesn’t have that demand. For instance, it would be theoretically possible (if it suited your purposes) to recognise a clade Sylvioidea with clades Hirundinidae and Aegithalidae within it, and treat all other sylvioids as unplaced below Sylvioidea.

  46. #47 Dartian
    February 9, 2011

    Before this thread turned into an Umpa-Lumper soapbox, Allen Hazen asked:

    Has anyone tried to estimate what the true number of species of, say, moles is?

    For the Western Palearctic Talpa genus, at least, there are some recent studies that relate to your question and that might interest you. Colangelo et al. (2010) provide molecular clock estimates of divergence dates within this clade. And Tryfonopoulos et al. (2010) and Canestrelli et al. (2010) show that there are high levels of intraspecific genetic variability within Talpa stankovici and Talpa romana, respectively; Tryfonopoulos et al. (2010) also demonstrate, by molecular methods, the previously unknown presence of Talpa stankovici in Greece.

    Canestrelli, D., Aloise, G., Cecchetti, S. & Nascetti, G. 2010. Birth of a hotspot of intraspecific genetic diversity: notes from the underground. Molecular Ecology 19, 5432–5451.

    Colangelo, P., Bannikova, A.A., Kryštufek, B., Lebedev, V.S., Annesi, F., Capanna, E. & Loy, A. 2010. Molecular systematics and evolutionary biogeography of the genus Talpa (Soricomorpha: Talpidae). Molecular Phylogenetics and Evolution 55, 372–380.

    Tryfonopoulos, G.A., Thanou, E.G., Fraguedakis-Tsolis, S.E. & Chondropoulos, B.P. 2010. New data on the distribution and genetic structure of Greek moles of the genus Talpa (Mammalia, Talpidae). Journal of Zoological Systematics and Evolutionary Research 48, 188–193.

  47. #48 John Harshman
    February 9, 2011

    if you want to break up passerines in the ‘lumpiest’ way possible, how about (1) acanthisittids, (2) eurylaimides, (3) tyrannides, (4) menuroids, (4) meliphagoids, (5) ‘core corvoids’, (6) petroicids, (7) passeroids, (8) stenostirids, (9) regulids, (10) ‘picathartoids’, (11) bombycillids, (12) parids, (13) muscicapoids, (14) certhioids and (15) sylvioids. This still leaves out quite a few taxa that are of uncertain placement.

    I realize this was intended purely as an exercise in absurdity, but there are many groups left out of this list, including many whose relationships we can be quite certain of, especially the long ladder of taxa making up “Corvida”, but not included in any monophyletic Menuroidea, Meliphagoidea, or Corvoidea.

    It should be mentioned again that there are objective criteria for taxon rank, but almost nobody has ever used them. And that’s with good reason, since they’re operationally difficult. I’m talking about genetic distance and/or age categories. Sibley & Ahlquist are the most obvious example. But they still had to split passerines into many, many families. Still, their “Fringillidae” is getting close to big enough to satisfy some of the lumpers here. Of course that’s only within birds. John Avise attempted something similar with all vertebrates, but if I recall even he quailed at the attempt to equate vertebrate families with insect families.

  48. #49 Vladimir Dinets
    February 9, 2011

    David: Thank you. I currently study the performing arts of crocodilians. Recently published a paper on dances, and the thesis is on songs.

    You make it sound like ranks are totally given up on, and only rankless systems are in use. This is far from being the case. Ranks are still being taught even to graduate students, they are widely discussed in papers in all areas, they are routinely used in, for example, paleontology to describe how well a certain group was doing at a certain time, and, above all, they are given a lot of attention in Wikipedia, which, like it or not, is rapidly developing into the official depositary of the summary knowledge of our species :-)

    By diversity here I mean presence of major differences in morphology, body plan, life history etc. within a group. In Mammalia, you get seriously different body plans, egg-laying vs. vivipary (of two kinds), animals as different as whales, naked molerats, bats and sloths. Birds all look pretty much alike, especially if plucked – their range of life forms is about the same as in large mammalian orders such as Carnivora or Rodentia. The only changes in life history whatsoever are the gradual shift towards naked chicks and the lack of brooding in megapodes.

  49. #50 Darren Naish
    February 9, 2011

    Well, let’s just hope there aren’t any ornithologists hanging around who might read this :)

  50. #51 Darren Naish
    February 9, 2011

    Oh, or non-Bentonite palaeontologists :)

  51. #52 Allen Hazen
    February 9, 2011

    Re: #47: Thanks, Dartian! (I suppose Canestrelli et al. just couldn’t resist the temptation to subtitle a paper about moles “Notes from the underground.”)

    Re #48: In principle one COULD define objective criteria for taxon rank, but — as you note — in more than one way. My guess is that, even within, say, Mammalia, the genetic difference criterion (which IS used by some authors to distinguish species from subspecies) and the age-since divergence criterion might lead to different rankings: different lineages can and sometimes do evolve at different rates.
    (And trying to use these criteria generally would probably mean scrapping FAR too many traditional assignments. In terms of age-since-divergence, don’t we sometimes have vertebrate orders younger than insect genera?)

  52. #53 William Miller
    February 10, 2011

    The problem with objective (time-divergence and genetic-difference) criteria for genus/family/etc are twofold:

    -actually defining them (5 million years divergence=genus, 15 million years=family, or whatever) is going to be arbitrary, though I suppose this wouldn’t really matter if the criteria were used by everybody (but how do you *get* that agreement?)

    -(and this is what Sibley/Ahlquist’s families ran into) they don’t reflect the ‘apparent’ groups very well.

    I personally think we ought to have and name clades (under something like the PhyloCode), AND have and name traditional ranks (under something like the ICZN). Ideally they have different purposes.

    Ranked taxa should be used as a convenient *nomenclatoral*, *organizational* system… named clades as a way to talk about the evolutionary units themselves. But saying the ranked taxa should be clades (eg the ‘genera must be monophyletic’ thing) is I think not ideal from either perspective. Linnaean ranks weren’t designed from a cladistic, or even an evolutionary, perspective, and I think we might as well accept that they don’t work so well for that.

    But they are still just as useful for their *original purpose* as they ever were (IIRC Linnaeus didn’t believe families and above were actually natural groups, just useful categories… not sure about genera.)

  53. #54 David Marjanović
    February 10, 2011

    You make it sound like ranks are totally given up on, and only rankless systems are in use. This is far from being the case. Ranks are still being taught even to graduate students, they are widely discussed in papers in all areas, they are routinely used in, for example, paleontology

    Most fields of vertebrate paleontology have, indeed, given up on ranks or pay them at most lip service.

    to describe how well a certain group was doing at a certain time,

    This is becoming rare, precisely because it’s misleading.

    and, above all, they are given a lot of attention in Wikipedia, which, like it or not, is rapidly developing into the official depositary of the summary knowledge of our species :-)

    Yes, and next time I log in I will lobby heavily against this. Have you seen how Wikipedia classifies languages?

    By diversity here I mean presence of major differences in morphology, body plan, life history etc. within a group. In Mammalia, you get seriously different body plans, egg-laying vs. vivipary (of two kinds), animals as different as whales, naked molerats, bats and sloths. Birds all look pretty much alike, especially if plucked – their range of life forms is about the same as in large mammalian orders such as Carnivora or Rodentia. The only changes in life history whatsoever are the gradual shift towards naked chicks and the lack of brooding in megapodes.

    How do you quantify that, why do you choose to emphasize life history, and… why do you compare Mammalia and Neornithes when these two clades have never been given the same rank? :-)

    IIRC Linnaeus didn’t believe families and above were actually natural groups, just useful categories… not sure about genera

    IIRC, he doubted even genera late in his life… He didn’t invent the rank of family, though. He placed great emphasis on there being exactly five ranks: kingdom, class, order, genus, species. He even claimed all classification systems of anything have five ranks, too.

  54. #55 Dartian
    February 10, 2011

    Allen:

    Thanks, Dartian!

    No problem. You might also want to check out:

    Sánchez-Villagra, M.R., Horovitz, I. & Motokawa, M. 2006. A comprehensive morphological analysis of talpid moles (Mammalia) phylogenetic relationships. Cladistics 22, 59–88.

  55. #56 Dartian
    February 10, 2011

    The only changes in life history whatsoever are the gradual shift towards naked chicks and the lack of brooding in megapodes.

    Cowbirds, whydahs, honeyguides, many cuckoos, and a few others don’t seem too keen on brooding either…

  56. #57 Vladimir Dinets
    February 10, 2011

    David: the day rankless systems will be the only ones left, the whole discussion of families and orders will become pointless, of course :-)

    Neornithes, Aves, what’s the difference in this case? I don’t think Archeopteryx had dorsal fins or live birth.
    If you don’t like life history, what would you prefer?

    Just name some area in which birds show broader spectrum of designs than a large mammalian order. Unless, of course, it’s some bird-specific thing like feather coloration.

    Dartian: OK, so the diversity of reproductive systems in Aves is about the same as in Apidae.

  57. #58 Dartian
    February 11, 2011

    Vladimir:

    OK, so the diversity of reproductive systems in Aves is about the same as in Apidae.

    I thought you were comparing birds to mammals? Please don’t move the goalposts.

  58. #59 Vladimir Dinets
    February 11, 2011

    Dartian: in mammals the diversity of reproductive systems is much higher anyway. In addition to the three main types, there are eusocial species, those that mate almost immediately after birth (some Mustela), those with lactation/parental care shortened to just 4 days (hooded seal), and those with 25+ years of parental care (H. sapiens in some populations).

  59. #60 David Marjanović
    February 11, 2011

    If you don’t like life history, what would you prefer?

    Instead of asking “why do you choose to emphasize life history”, I should have asked “why do you choose to deemphasize everything else” or “why do you choose anything at all, given that all such choices must be arbitrary”.

    It’s arbitrary. That’s what I don’t like. I want as much science as possible in my scientific nomenclature.

    Just name some area in which birds show broader spectrum of designs than a large mammalian order. Unless, of course, it’s some bird-specific thing like feather coloration.

    Skin colouration.

    Or indeed, why not compare feather colouration to hair colouration. Where are the green mammals?

  60. #61 Cameron
    February 11, 2011

    There have been some descriptions of “green” mammals , however the apparent greenish coloration is actually due to interference patterns

  61. #62 Vladimir Dinets
    February 11, 2011

    David: no, it’s not arbitrary. An animal is a bunch of morphological and functional (life history) characters… what else?

    Skin coloration in mammals is extremely variable: just look at the mandrill. But you can’t compare feather and hair coloration, because in so many cases feather colors are structural, and you can’t have it in hair. Besides, most mammals are nocturnal – I don’t remember any bright colors in owls or nightjars.

  62. #63 David Marjanović
    February 13, 2011

    David: no, it’s not arbitrary. An animal is a bunch of morphological and functional (life history) characters… what else?

    It’s arbitrary which characters you choose as “important”.

  63. #64 Vladimir Dinets
    February 13, 2011

    I chose morphology and life history. These are not characters, these are broad categories covering most of characters.

  64. #65 John Harshman
    February 14, 2011

    Morphology and life history cover most characters? Then what am I going to do with the billions of genetic characters?

  65. #66 farandfew
    February 15, 2011

    Is the question of whether there’s greater ‘disparity’ (is that word still allowed?) in mammals than birds really a question of classification or ranking? It seems a worthwhile scientific question in its own right and the comparison is justified by the similar species counts for the two clades, rather than by any conception of their being at a similar ‘level’ phylogenetically – whatever that might mean.
    These sorts of questions become important for prioritizing conservation as the idea of all species being equally important becomes less and less tenable – or less sellable anyway.

  66. #67 Dartian
    February 15, 2011

    farandfew:

    It seems a worthwhile scientific question in its own right

    Well, it is, but you can’t do science without quantifiable data. Which is what I (and others) have been trying to ask for in this thread. Thus far, only opinion and the occasional anecdote have been offered.

  67. #68 Vladimir Dinets
    February 15, 2011

    John: Genetic characters are pretty much meaningless for ranking – thy only show the sequence of divergence events. Differences in sequences can accumulate for millions of years without any major changes in the phenotype, but, on the other hand, the phenotype can change completely without corresponding differences in genetics. Just look at us and chimps – we should be congeneric if judged by molecular evidence alone, yet there are hundreds, if not thousands, of differences in anatomy and physiology, some of them on the level of differences between orders in birds.
    Dartian: of course you can. A lot of science is done without quantifiable data. I don’t remember seeing that many numbers in The Origin of Species, for example.

  68. #69 Bradley
    February 15, 2011

    Vladimir: Did I read that right. Did you say that humans and chimps are as different as orders of birds? This thread officially has become a real joke!

  69. #70 farandfew
    February 15, 2011

    I think the real problem is that science is very difficult without clearly defined concepts – quantifiable or not. In the case of systematics, the concepts have now been very well defined. Vladimir, if you feel that humans and chimps are more different than some bird orders, I don’t think it’s a joke because you haven’t defined what you mean by ‘more different’. Under some conceptions of what ‘more different’ means, you may well be right. But systematics is no longer based on an idea of what is more or less different. As David M says, it’s arbitrary which characters you choose as important.
    But that doesn’t mean that it *has* to be arbitrary, or that it’s impossible to define what ‘more different’ means. I just don’t think it can be done in the context of systematics or ‘ranking’. There may be several different systems we could use.

  70. #71 Vladimir Dinets
    February 16, 2011

    Bradley: humans have subcutaneous fat, more facial muscles, different throat anatomy, different dentition, different sweat composition, different feet anatomy, different ribcage shape, huge differences in female physiology and anatomy, different pelvic anatomy, and hundreds of other differences in morphology. I don’t think there are that many structural differences between, say, honeyguides and passerines.

  71. #72 Vladimir Dinets
    February 16, 2011

    farandfew: it’s true that if we use non-ranking system, it doesn’t matter how different two taxa are: all you need to know is the order of branching events. But if we want to assign rank, we need to know what qualifies for a family or an order.

  72. #73 Dartian
    February 16, 2011

    Vladimir:

    Genetic characters are pretty much meaningless for ranking

    That statement only makes sense if one thinks that ranking itself is meaningful (and quite a few regulars on this blog would disagree strongly with that).

    there are hundreds, if not thousands, of differences in anatomy and physiology

    Why do you only consider the differences between humans and chimpanzees and ignore the similarities? That sounds like splitting – not lumping – to me…

    I don’t remember seeing that many numbers in The Origin of Species

    Apart from the obvious fact that Origin of Species was written in the mid-nineteenth century, when science was done rather differently (e.g., there was no peer review back then), Darwin’s book is, and was meant to be, semi-popular. You can’t use it as a standard for today’s scientific discourse.

    (Without belittling Darwin’s immense and crucial contributions in any way, I’d say that evolutionary biology didn’t become a real, testable science until long after his death; it took Ronald Fisher, J.B.S. Haldane, Sewall Wright, and quantitative population genetics to do that. But that’s another story.)

  73. #74 David Marjanović
    February 16, 2011

    farandfew: it’s true that if we use non-ranking system, it doesn’t matter how different two taxa are: all you need to know is the order of branching events. But if we want to assign rank, we need to know what qualifies for a family or an order.

    So… do we want to assign rank?

    What for?

    Ranks are actively misleading, as we see every time again when someone writes a paper saying “of course ranks are subjective” and then immediately goes on to count orders, families or genera in order to estimate biodiversity. This happens all the time.

    quantitative population genetics

    is so mathematical that it needed math that didn’t even exist yet. Multivariate statistics was invented by population geneticists for population genetics.

  74. #75 Bradley
    February 16, 2011

    I do not understand Vlad’s type of argument. Yes, humans and chimps are different (but some of those differences you point to are incorrect actually). They are also alike. They are more alike than either is like gorillas, or orangs. In fact, I would go so far as to say that humans are chimps are really, really similar, those differences you list are of the sort you would expect between close genera.

  75. #76 Vladimir Dinets
    February 16, 2011

    Dartian and David: the benefits (or lack thereof) of ranking and non-ranking systems are a different subject. What I am saying is that ranking is applied to birds in ways inconsistent with its application to other vertebrates. If we decide to abandon ranking, the whole discussion would be totally meaningless, of course.

    Dartian: popular or not, Darwin’s work (and, by the way, Wallace didn’t use that many numbers either) was arguably the most influential publication in the history of science. May be more recent works rely more on numbers because they all deal with less fundamental issues, and nobody has the insight to come up with a theory as revolutionary as Darwin’s?

    Bradley: To find the same difference in physiology as between human women and chimp females, you’d probably have to compare birds with crocodiles. That’s exactly what I am talking about: what would be two genera in mammals or herps is considered two families, or even two orders, in birds.

  76. #77 farandfew
    February 17, 2011

    David Marjanović

    Ranks are actively misleading, as we see every time again when someone writes a paper saying “of course ranks are subjective” and then immediately goes on to count orders, families or genera in order to estimate biodiversity. This happens all the time.

    Two other phenomena about which similar arguments are made: GDP and IQ. Everyone complains about them but the fact is that they measure something we want to measure. They do so very imperfectly but imperfectly remains better than not at all.
    I want to know: Are ranks supposed to measure anything? If so, what? One thing or many things? Does it really need to be measured? Are ranks the best way to measure it/them?
    After these questions are answered, then we can *begin* to think about quantitative data. There is no point trying to ask for quantities until the qualities are agreed on.
    ‘Biodiversity’ is not a variable – as has been pointed out many times – it is not defined so clearl. It is often approximated for by species richness. Again, everyone complains about this. So when you say that it’s misleading to count orders, families or genera to estimate biodiversity, I think you really mean it’s a misleading estimate of species richness. But is that what we want to be measuring? Especially when a species is not what we thought it was when we first came up with the idea.

  77. #78 David Marjanović
    February 17, 2011

    Dartian and David: the benefits (or lack thereof) of ranking and non-ranking systems are a different subject. What I am saying is that ranking is applied to birds in ways inconsistent with its application to other vertebrates.

    What I’m saying is that ranking isn’t applied consistently to anything, and that it isn’t possible to apply it consistently to anything. This directly leads to an argument for abandoning ranks.

    Are ranks supposed to measure anything?

    Nope.

    Linnaeus believed they existed in the mind of God and he was discovering them — but even he wasn’t sure about the higher ranks.

    OK, the rank of species has been defined (in 147 incompatible ways as of February 2009). But all other ranks aren’t even supposed to measure something that exists in nature.

    So when you say that it’s misleading to count orders, families or genera to estimate biodiversity, I think you really mean it’s a misleading estimate of species richness. But is that what we want to be measuring?

    Probably not (and certainly not as long as we haven’t agreed on one species concept); that’s why the Phylogenetic Diversity Index was invented.

  78. #79 farandfew
    February 24, 2011

    I seem to have started a habit of getting into interesting discussions in the comments here and then disappearing.

    David M, it seems to me that you regard the maintenance of ranks and of vague phenetic species definitions as purely a result of inertia. You may be right – certainly it must be a very important factor – but I’m not so sure that’s all there is to it. However Linnaeus may have justified his system in the first place it has since been put to different uses. To look at the species question rather than the ranking one, is it desirable that we should all agree on a single species concept when the idea is used for such different things. The ‘species’ is the name we give to the taxonomic units that we care about (how often are phylogenetic diversity indices used to make real conservation decisions?). This is undoubtedly hard to measure and seems soft and sloppy but the dangers of not trying to measure it are very real. For example, someone is surely going to argue that the US Endangered Species Act is no longer valid because taxonomists have moved the goalposts. Politicians thought they were signing off on protection of morphologically or ecologically distinct populations. Now they’re unfairly tied into protecting things that are only different in terms of a handful of base pairs and this is justified in terms of…evolution!
    (well, sort of)

    And so with ranks. Perhaps the vague human perception about how much difference warrants a new rank is actually the best measure available of how distinct we feel something is. Perhaps that is something worth measuring in its own right. An idea which exists in the mind of man only but one which has power to affect which combinations of characters are going to actually be found on this planet in future.

  79. #80 David Marjanović
    February 25, 2011

    is it desirable that we should all agree on a single species concept when the idea is used for such different things

    Different species concepts describe different kinds of entities that have nothing in common except the word “species”.

    Some of these entities are very interesting for biology, so naming them isn’t a bad idea. But this would mean we’d end up with several overlapping nomenclatures.

    how often are phylogenetic diversity indices used to make real conservation decisions?

    Nowhere near as often as they should be*. Inertia again.

    * Well, if we already talk about “ought” instead of about “is”, such decisions shouldn’t be made at all. There’s way more than enough money available; what’s lacking is political will.

    For example, someone is surely going to argue that the US Endangered Species Act is no longer valid because taxonomists have moved the goalposts.

    There are already people who “argue” that the US Endangered Species Act never was valid, because Jesus is coming back any minute now!

    And there are already people who believe we cannot run out of natural resources because God would feel obliged to simply create more. There are a few top managers of big corporations who are this crazy.

    Perhaps the vague human perception about how much difference warrants a new rank is actually the best measure available of how distinct we feel something is.

    Who is “we”? Have you ever found two classifications (made by different experts, or the same expert at different times) that were identical? Ever?

    The complete lack of consensus, and the complete lack of a possibility of consensus, are two of the biggest disadvantages of rank-based nomenclature.

    has power to affect which combinations of characters are going to actually be found on this planet in future

    The science of phylogenetics has that power. Nomenclature only has it when it’s modeled to represent a phylogenetic hypothesis.

  80. #81 farandfew
    February 25, 2011

    The science of phylogenetics has that power.

    How can any science have that power? Except perhaps through inspiring people to act. Does phylogenetics? Sometimes – the EDGE programme is the best example I know but that’s not because of the nomenclature, just the trees.

    There’s way more than enough money available; what’s lacking is political will.

    Enough money available to save what, for whom?
    Conservationists have been trying for decades to raise political will to save ‘species’. It is extremely difficult. You don’t generally get much political will by just telling people what they should care about. People may argue that the Endangered Species Act makes no sense because Jesus is coming any minute but they don’t actually argue that very successfully. More successfully than one might wish, perhaps, but I doubt it’s a real danger at the moment. However the argument about moving of goalposts is not only dangerous but also actually justified.

    So when you say this would mean we’d end up with several overlapping nomenclatures. I want to make a claim for the original Linnean one. Keep ‘species’ to mean a vague, subjectively and inconsistently defined rank which society has already bought in to. Use some other word or words for the clades defined by more precise concepts. Taxonomists know that the naming is arbitrary, and biologists know that there are different concepts which can be used for classification. But the general public (including birders) and political leaders do not know this and, unfortunately for us, this matters. For this reason, the wholesale adoption of new species concepts will have a very real cost – one which we cannot even agree how to measure.

    The fact that two different experts won’t come up with an identical classification misses the point; an imprecise measurement of the right thing is better than a precise measure of the wrong one.

    Perhaps there is a better solution but I don’t know what it is.

  81. #83 David Marjanović
    March 3, 2011

    How can any science have that power?

    Sorry, I must have misread.

    Enough money available to save what, for whom?

    Everything for everyone.

    Conservationists have been trying for decades to raise political will to save ‘species’. It is extremely difficult.

    That doesn’t contradict what I say. I say the only problem is the lack of political will; money is not an issue, political will is.

    http://imgs.xkcd.com/comics/herpetology.png

    http://scienceblogs.com/pharyngula/2011/03/soon_to_appear_on_a_thousand_h.php

    Also, it is evil to link to an isolated picture from xkcd. You need to link to the page the picture is on, so people can see the alt-text. The alt-text is a necessary part of the composition.

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.