PBS: How many species of giraffes are there?

A while back I discussed a paper in Current Biology by Julian Fennessy et al. . That paper used genetic analysis (the total genetic divergence among groups) to claim that there are actually four species of giraffe instead of a single species with nine subspecies. Using the Biological Species Concept (BSC), however, I argued that there was no objective basis for recognizing four distinct species on the basis of genetic distance and monophyly alone, for such recognition is purely subjective. How much genetic divergence between geographically isolated groups is necessary before we call them “separate species”? Any decision must necessarily be subjective, since no cut-off point of genetic distance is biologically meaningful.

I concluded that although the press gave the Fennessy et al. paper a ton of publicity, there’s no good reason to recognize four instead of one species of giraffe so long as all the “species” are geographically isolated from one another. (Greg Mayer and Matthew Cobb, my biology co-writers here, agreed.)

Now, mirabile dictu, the Public Broadcasting Service (PBS) in the U.S. has taken up the issue, and I had several conversations about speciation with writer Becca Cudmore, who proved to be one of the more inquisitive and savvy science journalists I’ve encountered. And, miracle of miracles again, she gives substantial publicity to the idea that giraffes may not really comprise four species.

In her PBS NatureNow article “How many giraffe species are there, really?” Cudmore gives a good airing of the BSC and my take on the giraffes. The gist:

Unlike Coyne’s approach, the study used genetic differences to separate the giraffes. This is a method of defining species by their “phylogenetics,”or by their shared traits. In this case, Goethe University researcher Axel Janke found genetic markers, such as mutations, that were common among certain giraffes and not shared by the others. This suggested to him that there has been very little gene sharing between the groups.

But by Coyne’s definition, this doesn’t prove that giraffes are reproductively isolated. “The only way to show whether or not they are separate would be to move the wild giraffes into the same area and see if they produce a fertile offspring,” he says. While the different subspecies are known to hybridize in captivity, there is very little evidence of this in the wild.

“The Biological Species Concept is more meaningful because it helps to explain one of evolutionary biology’s most profound questions”: Why is nature discontinuous? he says—why is it not all “one big smear” that can exchange genes?

And so on. There is of course some pushback:

Still, with what we know about hybrids between species in the wild, Janke calls Coyne’s approach too “pure” and says that it’s going out-of-date.

Janke is just wrong here. (I have no idea what he means by “too pure”!) The fact that some species exchange genes is not a serious problem for a concept based on reproductive isolation between entire genomes, and in fact most closely related species do not exchange genes. The cases of gene exchange between biological species, while widely publicized, are not the rule but the exception. (See Coyne & Orr, Speciation, for the evidence.)

Most tellingly, virtually every paper I’ve seen on the process of speciation—that is, on the ways that new species come into being—deals not with the accumulation of genetic distance per se, but on the development of reproductive barriers that eventually prevent populations from exchanging genes. That’s a tacit admission of the importance of the BSC.

I think the impetus behind naming more giraffe species is largely connected with conservation, for with more named species we can put more species on the endangered list and save more of the phenotypic and genetic diversity in what was formerly one species. But while that may be an admirable goal, it should not be a motivation for recognizing species in nature.

It may not be a coincidence that Fennessy works for the Giraffe Conservation Foundation. Cudmore notes this:

Whether one, four, or six species, giraffes have experienced a 40 percent plummet in population over the past 15 years. They’re currently listed as a species of “least concern” by the IUCN [International Union for Conservation of Nature] and unlike alarm bells ringing for Africa’s elephants, gorillas, and rhinos amid the poaching crisis, they receive relatively little attention.

and the press release for the paper gives a quote from Fennessy:

“With now four distinct species, the conservation status of each of these can be better defined and in turn added to the IUCN Red List,” said study co-author Julian Fennessy of the Giraffe Conservation Foundation in a release. For example, said Fennessy, there are now less than 4,750 Northern giraffes and fewer than 8,700 reticulated giraffes in the wild. “As distinct species, [this] makes them some of the most endangered large mammals in the world.”

This makes me suspect that behind the “splitting” of giraffes is a conservationist motivation, not an attempt to partition out nature in biologically and evolutionarily meaningful ways.


(from PBS article) A recent study proposed that giraffes are actually comprised of four main species (from left to right): reticulated, northern, southern and Masaai.


  1. Klendathu
    Posted November 2, 2016 at 8:39 am | Permalink

    How can those be different species while Asians and Aboriginals are the same.

    Blows my mind.

    • eric
      Posted November 2, 2016 at 9:32 am | Permalink

      Not every genetic difference is necessarily apparent. Its possible that the genetic differences between the four giraffe groups is larger than the differences between human groups, even if visually the giraffe groups look a lot more similar to each other than the human groups. I don’t know whether that’s actually the case, I’m talking more ‘in principle.’

      You also have to consider the possibility of cognitive bias. Some amount of our brains is specialized in human facial recognition. Presumably, giraffes don’t have this but they may have brain capacity specialized in seeing differences between giraffes. That means we see and process much finer differences between human individuals than between giraffe individuals, and vice versa. Ask a (hypothetical sentient) giraffe if the four giraffe groups look ‘more different’ than the various human groups, and they might very well say ‘yes, it blows my mind some human could think otherwise.’ 🙂

      • TJR
        Posted November 2, 2016 at 11:19 am | Permalink

        Yeah, those giraffes all look the same to me. Let’s build a wall to keep them out, and make them pay for it.

  2. John Harshman
    Posted November 2, 2016 at 9:14 am | Permalink

    Well, there are multiple species concepts, as I believe you note in your book. The phylogenetic species concept, in particular, considers significant genetic divergence or private alleles as warranting species status. And if left alone for long enough, phylogenetic species will eventually become biological species. So perhaps all we have to do is let this controversy go for a few tens or hundreds of thousands of years, and it will all become moot.

    • John Harshman
      Posted November 2, 2016 at 9:16 am | Permalink

      You overestimate the divergence between human groups. The cutoff is subjective, but if there is one, there’s a lot of room to put it in a place that makes for separate giraffe species but not separate human species.

      • John Harshman
        Posted November 2, 2016 at 9:17 am | Permalink

        That was supposed to be a reply to Klendathu.

  3. µ
    Posted November 2, 2016 at 9:22 am | Permalink

    The “too pure” approach must refer to Jedi mindfulness when partitioning biological diversity: Are there pre-mating or post-mating reproductive barriers, or not?

    It seems to take superhuman Jedi powers to think clearly about all that, hence “too pure”.

  4. Achrachno
    Posted November 2, 2016 at 9:51 am | Permalink

    “Any decision must necessarily be subjective, since no cut-off point of genetic distance is biologically meaningful.”

    Trouble is, in practice it’s often subjective with the BSC too — one seldom has nice clean species that are sympatric, never cross — and where one has good evidence showing how and why that’s true.

    In my experience (plants) when doing a revision of a large group of taxa (a genus, say) what one finds is some tentative taxa showing no signs of introgression, some showing a little here and there, some with none over most of their range but extensive in one little area, some mostly distinct but apparently “leaking” fairly regularly, and some are morphologically distinct, maybe in fairly large ways over vast areas, but with an extensive zone of hybridization in significant part of their range. Plus other messy variations. There is no clear cut-off point in reproductive isolation. Clearly the existence of a lone hybrid is not grounds for merging two species, but what about 2?, 47? and so on.

    Plus, there are lots of taxa that are geographically isolated, morphologically divergent, and perhaps diverge in other ways (ecological/physiological) too. Often they diverge in ways and to a degree that is similar to divergence in related taxa that are known to be good species under the BSC, but one does not have all the needed information in the taxon at hand. Many of the species I work with are known only from dead preserved specimens — and sometimes only from one or two old (no living person has ever seen the plant alive) specimens. But, the difference may be so great that I can’t believe this old specimen is the same as some better known species. And, it will often have been described as a full species long ago, and I’m usually going to go along.

    Personally, I’m OK with what the giraffe workers have done. They seem to have more data than most of us have most of the time.

    • Posted November 2, 2016 at 11:08 am | Permalink

      I too work in plants, and have the same experience as you. Plants are far messier than vertebrates, and standard species concepts which work well for animals may not work for plants. I generally treat the biological species concept as the gold standard–when multiple forms are sympatric and don’t interbreed, these are good species. I then try to look at the kinds of morphological differences those forms exhibit, and if I see other forms that are not sympatric, but similarly divergent, I will cautiously treat them as new. The degree of morphological differtiation that leads to reproductive isolation in sympatry vary considerably between genera and tribes. In some groups (eg Gongora orchids) some species are visually nearly identical (to our eyes) but are separated by fragrance.

  5. Posted November 2, 2016 at 9:59 am | Permalink

    Is there really a difference between not inter breading because it’s biologically impossible and not inter breading because it’s geographically impossible?

    The net result is the two subspecies will continue to diverge anyway because genes are not being mixed.

    It strikes me as a technicality that’s not worth the fight.

    • Achrachno
      Posted November 2, 2016 at 10:20 pm | Permalink

      Yes, in a sense geography is just another isolating mechanism: one of the quickest and most effective. A single long-distance dispersal can create a new evolutionary lineage, or several. The Hawaiian silverswords and relatives (c. 28 species in 3 weak genera) are the product of one long dispersal from California — and it was the dispersal that was effectively the key speciation event. Once isolation happened, evolutionary divergence was inevitable.

  6. Dominic
    Posted November 2, 2016 at 10:09 am | Permalink

    This is relevant –
    Trait biogeography of marine copepods – an analysis across scales
    Predictable convergence in hemoglobin function has unpredictable molecular underpinnings
    Fine-scale behavioural differences distinguish resource use by ecomorphs in a closed ecosystem

    If PCC[E] has not read them he may like to – I am sure he will have a view!

  7. W.Benson
    Posted November 2, 2016 at 10:54 am | Permalink

    More than the question of being valid species or not, an important point is that there are numerous giraffe populations that are genetically and possibly ecologically (were they studied adequately)distinct that merit protection in wildlife reserves. Another issue, minor I hope, is that taxonomists seem sometimes willing to stretch the rules on what is a species to get their names on new species: Undescribed species of mammals and birds are running out as is the profession of museum taxonomy.

    • Posted November 2, 2016 at 11:10 am | Permalink

      There are still plenty of plants and bugs to name after people!

      • Graham
        Posted November 2, 2016 at 12:27 pm | Permalink

        And fish – hundreds of undescribed species of catfish alone.

  8. Alpha Neil
    Posted November 2, 2016 at 11:19 am | Permalink

    “The only way to show whether or not they are separate would be to move the wild giraffes into the same area and see if they produce a fertile offspring,”

    Creationist website headline: “They have no shame: Evolutionist proposes taxpayer-funded Giraffe orgy”

    • Posted November 3, 2016 at 2:48 pm | Permalink

      When I read about human genetics, I sometimes imagine a dystopia where a geneticist tyrant tells individuals with this or that trait whom to marry. (And may also put their progeny in standardized environment.)

  9. Posted November 2, 2016 at 12:11 pm | Permalink

    A thought from the cheap-seats. It seems that Jerry is doing *basic science*, and the conservation biologists are doing *applied science* or maybe *technology*.

    Maybe there are species concepts that *do* make better sense in biotechnology rather than bioscience.

    It seems to me that one of conservation biology’s interesting questions is *what is to be conserved*, and that might require different notions.

  10. Diane G.
    Posted November 2, 2016 at 6:59 pm | Permalink

    Subbing to support the science posts.

    Also, a thought–would it be possible that splitting into four species might be detrimental to conservation by increasing trophy hunting by those wanting to have one specimen of each species?

    • Achrachno
      Posted November 2, 2016 at 7:44 pm | Permalink

      Only if trophy hunting is a major threat, which I’ll bet it’s not. The bush meat trade and land conversion are probably bigger problems. Just my guess.

      • Diane G.
        Posted November 2, 2016 at 9:10 pm | Permalink

        Makes sense.

        • keith cook +/-
          Posted November 3, 2016 at 2:53 pm | Permalink

          I just happen to see a disturbing programme of a protection group investigating a 4 giraffe shooting. A rare sight in this particular park these animals were not shot for meat, just for the very end part of the tails (the hairy bit) for ceremonial purposes. Trying to remember, but i think it was to do with matrimony.. dowry perhaps?

  11. FloM
    Posted November 3, 2016 at 7:52 am | Permalink

    “species of giraffe”, not “species of giraffes”

%d bloggers like this: