on lemur systematics

It looks like the free-for-all species splitting banquet is coming to an end. Ian Tattersall, the  leading expert on Madagascar lemurs (one of sifaka species is named after him) has raised his voice against the ridiculous splitfest, and proposed synonymizing 50% of all lemur species. I hope the same will be done for other primates, because they are the most ridiculously oversplit group at the moment. http://www.ajol.info/index.php/mcd/article/view/91494/80973  


  • John Fox

    I don’t know, V, I kind of like the splitters. The whole species concept breaks down at a certain point, likely where speciation is occurring. If the species concept has no meaning, why not point out the differences between similar animals?

    In other words, the lumpers have no more justification for their point of view than the splitters, but the splitters give life to the differences and illuminate the issues. The splitters create a record that can be examined in the future. If the question is just one of semantics, what harm is done?

    Besides, I hate taking critters off my life list, lol.

  • vdinets

    John: there are no real lumpers today, just like there is no real left in US politics. There are just people who try to remind others that splitting should be checked by some criteria. Otherwise species lists become meaningless. Consistent application of “Phylogenetic Species Concept” would lead to splitting Peromyscus maniculatus into about three thousand species, and Homo sapiens into a few hundred. And much of the splitting is simply based on flawed procedures, like relying entirely on mtDNA differences or vocalizations.
    Note also that species lists are not just menus for twitchers. They are used for setting conservation priorities, for estimating biodiversity levels and rates of evolution, for paleoecological studies, et cetera. For example, splitting lemurs into 120 species makes it look like Madagascar forests have been severely fragmented for millions of years, while in reality there was continuous forest cover over much of the island until just a few centuries ago, and many of those “species” are either fragments of former clines or variants recently created in small forest fragments by founder effect. If a reforestation program gets implemented, most of these “species” will happily merge again.

    • Don Roberson

      I whole-heartedly agree with Vlad on this. The last 2 volumes of “Handbook of Mammals of the World” is very disappointing with oversplit taxa. The line has been held pretty well within the bird world — with which I am much more familiar — with Biological Species Concepts remaining primary and at least logically satisfying — so I am glad to hear that someone is attempting to stop the slippery PSC slide in the mammal world.

      • vdinets

        Thanks for the support! But birds… have you seen the recent description of a new tailorbird from Cambodia? The authors were so afraid that if they compared the new “species” with allopatric birds from the old species there would be zero genetic difference, they went to the trouble of procuring the old species’ specimens from Indonesia and using them in the analysis (didn’t help much, but they published it anyway).

      • morganchurchill

        My views are probably well known here, but while BSC works well in birds…it’s not terribly useful in other groups. I have personally recommended in the literature the lumping of taxa in some groups. That said mammals overall are probably overlumped.

        Remember that species are not absolute platonic concepts. Where you draw the line regardless of species concepts is an arbitrary decision. I happen to believe that the criteria of interbreeding is extraordinarly difficult to test in many groups, especially given the prevalence of allopatry, the obscurity of species recognition systems in nonbird groups, and frequency interbreeding can occur in very disparate groups (see ducks, snakes, etc)

      • vdinets

        morganchurchill: BSC has evolved past the simple “interbreeding or not” criteria decades ago. The most recent version is called “Differential Fitness Species Concept”. Have a look at: Hausdorf, B. 2011. Progress toward a general species concept. Evolution 65, 4: 923–931. (doi:10.1111/j.1558-5646.2011.01231.x)

  • morganchurchill

    Which still doesn’t get around the problems I state above, nor does it address to what degree and what traits are “negative” when interchanged (especially since hybrid speciation can obscure the latter, and traits evolve alongside environments and communities).

    Every few years there are flurries of papers that argue one approach over the other, or attempt to improve existing concepts. To the point where I think something like 100 species concepts have been named. I don’t think we are ever going to get consensus or acceptance of a single concept in zoology.

    • vdinets

      I agree that no species concept will ever be perfect or even universal, in part because species evolve and interact differently in different major groups. But PSC, being non-falcifiable, is purely unscientific and should never be used.

      • morganchurchill

        How is it non falcifiable? and how do you falsify BSC for allopatric populations that never come into contact with each other in nature?

        With PSC you can potentially set up some benchmarks, such as monophyly and degree of genetic or morphological distinctiveness.

        In BSC, your left with either defining some sort of hybrid zone size or frequency of interbreeding, or setting some benchmark of how distinctive an organisms vocalizations/morphology must be to be a species. Same problem in that ultimately you have to pick an arbitrary cut off point.

        • vdinets

          It is non-falsifiable because any difference between populations, no matter how small, can fulfill the “diagnosticability” criteria. For example, if you can tell a Tutsi from a Hutu (their facial features and stature are remarkably different, and so are their ways of life, plus there is obvious assortative mating) you can split them as PSC species. Monophyly comes pretty much automatically for almost any somewhat isolated population. As for the degree of genetic or morphological distinctiveness, people get around it all the time by a variety of methods. They can arbitrarily choose one over the other, for example, or they can use shifting baselines. For example: golden-cheeked warbler is a poorly differentiated species that should never have been split in the first place. Its small genetic difference from black-throated green warbler has been used as a justification to split other “species”, claiming that “they are as distinct genetically as some traditionally recognized warbler species”. Then the fact that some of these new “species” look almost identical can be used for new splits without referring to molecular data; these new, genetically nearly-identical splits can be used to justify yet new splits on tiniest molecular differences, and so on, ad infinitum. Or you can simply use small sample size and claim individual differences to be universal, like with splitting the golden palm-civet based on (if I remember correctly) 3 or 5 skins. Or you can use just one feature, like mtDNA or vocalization, and assign absolute importance to it. The result is a chain reaction of never-ending splitting.

  • morganchurchill

    You are using an extreme view to dismiss PSC. Which is why you need benchmarks. Humans have morphological differences between populations, sure, but genetically we are not terribly diverse. That is why you need to sample across a broad array of genes and create some sort of benchmark of differentiation (which needs to be determined for each group separately. And pair these results with rigorous morphologic analysis.

    I mean I could easily argue the opposite with an extreme view of BSC. Their have been known intrageneric hybrids in whales…should we lump False killers in with Bottlenose? Or Southern Right Whale Dolphins with Hectors? Should we just have one species of Anas duck? One species of otariid seal?

    • vdinets

      Not really, I am not. Diagnosticability-based PSC is the most frequently used version nowadays. It’s true that humans are not terribly diverse genetically, but the differences between populations are very consistent, and behavioral reproductive isolation can be really impressive (Gypsies still look Indian after many centuries spent in the West). In fact, the case for splitting us into a few dozen species would be better justified than many recent primate splits. Note that many of recently split primates differ only in facial features, just as many human ethnic groups. Geographic variability of faces seems to be a typical primate peculiarity.

      On the other hand, it’s been a long time since anyone proposed lumping species based just on occasional hybridization. Even the most hardcore BSC proponents have accepted the existence of things like narrow hybridization zones, limited gene flows, even hybrid swarms.

      • morganchurchill

        Your last paragraph is what I am getting at. You are arguing for a nuanced version of the BSC and then using it to argue against a strawman PSC. There are researchers out there who apply the PSC willy nilly with weak science. But that’s bad science, not something shared by all PSC advocates. There has also been a lot of bad BSC, where groups of species were synomized with practically no reason given

        Almost all good studies I have seen using PSC usually require a concordance of morphological and genetic differences. Morphological comparisons are not sufficient without a strongly supported robust phylogeny, something you would have a problem with if applying to humans. There should be a rigor applied to taxonomy, regardless of which method you advocate. If you want to argue against bad taxonomy, I am with you. If you think bad taxonomy is the only possible consequence of PSC, then we have radically different and irreconcilable perspectives on science.

      • vdinets

        I am comparing the most current version of BSC with the most current version of PSC. While BSC has been evolving towards more reasonable criteria, PSC has been evolving towards more lousy ones. And even in its most strict version, PSC is still formulated so badly that anything can be split. It is a fundamentally flawed concept (see Zachos & Lovari 2013, http://www.italian-journal-of-mammalogy.it/article/view/8849/0). It is even less applicable to uniparental species than BSC (see Coyne and Orr 2004, http://www.joelvelasco.net/teaching/2890/coyneorr04-speciationch1.pdf), and, contrary to its proponents’ claims, it is bad for conservation (Zachos 2013, http://www.nature.com/nature/journal/v494/n7435/full/494035c.html).

        Note that there is, in fact, phylogenetic base for splitting Homo sapiens: northern populations have hybridized with H. neanderthalensis, while those in Oceania have also hybridized with Denisovians. Also, the out-of-Africa clade and many groups within Africa are strongly monophyletic.

  • morganchurchill

    Yeah your argument keeps circulating back to the same points…If you want to criticize studies for sampling and methodology issues, fine, I am right there with you. But you seem to conflate PSC as HAVING TO HAVE THESE PROBLEMS, while again arguing for nuances in the BSC. As for conservation, any issues can be alleviated by using evolutionary distinctiveness as a criteria for conservation effort…of which several studies have been published and which gets around the problem associated with conservation (which seem to be less the taxonomy is bad, so much as funding as limited…which seems a tad poor reason to advocate against a scientific methodology.

    I will note that people against PSC keep circulating back to humans, but will also note that a quick googlescholar search fails to reveal any relevant studies that actually split humans into multiple species. Again…Strawman

    • vdinets

      Yes, I keep circulating back to the same points because these points are never countered. BSC is getting more nuanced over time, while PSC is becoming more and more extreme without ever adressing its original flaws.

      Conservation issues were originally brought into the discussion by PSC proponents who argued that over-splitting is good for conservation. You are absolutely correct in pointing out that this is not true, but this argument is still used.

      No, there were no actual attempts to split humans, not since the original description (Linnaeus described 4 subspecies). And it’s unlikely to happen because the issue is so politicized. But if PSC was applied consistently, H. sapiens would have to be split into hundreds of species. The reason this is brought up is that we know more about our own species than about any other, so it’s a nice test subject for various species concepts. If any concept, being applied to H. sapiens, would result in such an insane splitting, the concept must be flawed.

  • morganchurchill

    and again…you continue to ignore my earlier comments. So I guess I am done here

  • Bob Berghaier


    An interesting exchange between you.
    However have you thought about the “conservation value” of splitting?

  • Bob Berghaier

    Thank you for your reply. I am aware of the scientific reservations for over-splitting. However on occasion over splitting for primates can give governments the incentive to set aside conservation areas that not only protect the primate species in questions but with the umbrella effect numerous other life forms as well.

    • Vladimir Dinets

      Everybody says that, but I am not aware of a single case where splitting “for conservation priority” actually worked. Most “new species” are described from already existing protected areas, and most countries have legal provisions for protecting subspecies as well as species. What “splitting for conservation” actually does is undermining the public trust in scientific basis for conservation. Remember, the people fighting against conservation measures are not all stupid, and they are ready to exploit any flaw in scientific justification. There was already a case in the US where anti-conservation lobby challenged the validity of one endangered, but poorly defined subspecies of jumping mouse in court. I’m sure we’ll see more such challenges in the future.

  • Bob Berghaier

    Vladimir, how about these three;

    Cross River gorilla – separate species or just subspecies of Western lowland? The Cross River region is exceptional for bio diversity and under a great deal of threat

    Lac Alaotra bamboo lemur (Hapelemur alaofrensis) is it really a full
    species? Lake Alaotra is very important habitat for endangered species on Malagasy waterfowl.

    Pennants red colobus (Procolobus Pennantii) a species I have worked on Bioko. Bioko subspecies maybe a full species different from the Niger Delta form in Nigeria. If both were considered separate species both could easily be listed as one of the top 25 endangered primates on the planet.

    If I looked further into doc and other Indochinese langurs similar examples could be found.

    While I agree that unscientific species splitting is a bad idea in the USA due to the fact that nearly any endangered species classification is controversial, since it is considered anti business, that does not necessary hold true for developing countries.

    • Vladimir Dinets

      Is there any evidence that classifying these three taxa as full species rather than subspecies would really change the level of conservation effort? Mountain gorilla has always been considered a subspecies, yet it is one of the most impressive conservation success stories in Africa. The effort to save Lake Alaotra bamboo lemur started well before it was claimed to be a full species. And the list of the world’s 25 most endangered primates includes subspecies as well as species. Bioko red colobus is already on that list (http://en.wikipedia.org/wiki/The_World's_25_Most_Endangered_Primates).

  • Bob Berghaier


    Thanks for the reply and your thoughts. There was a big push at the recent International Primate Conference in Mexico to add the Niger Delta red colobus to list that as a separate species in place of the Bioko form but it was decided that having two different species of red colobus on the list was “politically’ not a wise move and the Bioko red colobus was chosen again.

    Again I don’t see the harm in granting full species status to an endangered subspecies in a developing country. I’ll turn the discussion back to you. Are you aware of a backlash in developing country similar to the Pueblo jumping mouse which I assume is the USA case that you have referred to?

    • Vladimir Dinets

      Well, if both were still considered the same species, they could be added to the list jointly. Now that they are split, you have to choose one or the other. No harm done, you said?

      Yes, indeed I am. There is a long history of commercial butterfly collectors splitting subspecies into full species in order to avoid the procedure of applying for collecting or exportation permits. You see, if the old species was listed in the local Red Data Book or the CITES list, splitting off a subspecies as a new species automatically makes it unlisted in the eyes of local bureaucrats. That happened to a few Parnassius subspecies in Central Asia, for example.

      • Jon Hall


        Vladimir, a good reply. I was not aware of the butterfly issue. As for the red colobus choice, I was told by one of the participants of the committee that they wanted to chose the Niger Delta type as a separate species to highlight a critically endangered ecosystem. However they were overruled. I agree with you on the butterfly issue but as for primates I will stand by my position. As with many things in life “one size or one position” do not fit all conservation situations. Bob

Leave a Reply