Hi Tommy

Species is a man-made concept to ease classification. And there is several definitions of it (biological species concept, etc...).. There is no way molecular markers can tell you whether two individuals belong to the same of different species. Instead, once you have defined your species, you can try to find a molecular marker to support your previous classification. But even if you find some allele is present in one (or several) populations but not in others, you can only say there is genetic structure. Whether that genetic structure represents speciation is something you should decide based on your prior definition of species.

I do not know these particular studies, but generally there is nowadays the unfortunate tendency to substitute "modern" molecular data for serious (with the - only decisive! - study of presence/absence of intermediates, i.e. occurrence or lack of hybridization, in the border areas) taxonomic analysis! To me, taxonomic conclusions based only on molecular markers are simply unscientific. 

Nowadays we know that every author has its own position :)

... And you will find some people loooking and looooooooooooking to all genes of their dataset until they find THAT marker (in the middle of a neutral global result) that will show that HIS/HER SPECIES is statistically "real" (for example, frequencies of an allele coding for color under a strongly selective environment for that color). And the worse part: they get published in more-than-good journals. Now with RADs, for example, you could "find" a species almost everywhere, if you look enough.

It appears to me that the paper should not have passed the peer review with these conclusions.

It leaves an impression that they were very eager to draw much stronger conclusions than their data actually support. Even though their results might be valid and quite interesting.

There is a comment section below the original publication at cell.com, you can keep track of what other scientists are saying.

Thanks to all for your answers.

Roman, Mikhail, Leocadio, I agree that taxonomic conclusions will vary from an author to another, and that looking deeper and deeper you will often end finding subdivisions. It is an interesting debate, but more philosophical than my down-to-earth questioning that was about comparing results from two studies. Apart from the fact the 2016 paper included tissues from the nominate subspecies, the rest of it seems at the very least entirely redundant with the 2007 study, and at worst poorer (in terms of sampling census, lack of contact zones, lower resolution obtained with 7 vs 14 microsatellites loci). Thus my very question was: what (the hell) is new in the 2016 results? Not much, except boldness in conclusions? :)

Ivan, thanks for your comment, I kind of shared this feeling about the conclusions, but as I read about this discovery in Nature News & Comments (and later on, in 10 generalist newspapers), I tended to put in doubt my understanding first! ^^

I will keep track of feedbacks about this publication.

The only "real value" of the article that they tested only samples form WC stock. That's provide true data. And the conclusions as we all know depend on authorship. Anyone may conclude the opposite to others on the same data...

And sure, it is depends on peer-review as well.

Hi Tommy

Species is a man-made concept to ease classification. And there is several definitions of it (biological species concept, etc...).. There is no way molecular markers can tell you whether two individuals belong to the same of different species. Instead, once you have defined your species, you can try to find a molecular marker to support your previous classification. But even if you find some allele is present in one (or several) populations but not in others, you can only say there is genetic structure. Whether that genetic structure represents speciation is something you should decide based on your prior definition of species.

A response to Fennessy and colleagues has been published (20 Feb 2017) in Current Biology (http://dx.doi.org/10.1016/j.cub.2016.12.039). The second and third point is particularily interesting: it all happens as if Fennessy waves Brown's results away with the flick of a hand - or mostly ignores them - to make room for their own shaky conclusions.

https://10.1016/j.cub.2016.12.039

Please note that there is a morphologicall revision of Ungulates by Groves and Grubb (2011) that recognise more than four species in Giraffa adopting the phylogenetic species concept. This book is not cited by Fennesy et al.

As you can see from some of my publications, giraffes are a classical case in which a polythipic species obscures the grim conservation status of some of the components. So a multi-species classification - based on sound data - is certainly positive for conservation planning.

Gippoliti Spartaco unfortunately Groves and Grubb had a very small sample to work with. Some species (antiquorum, giraffa, camelopardalis) were considered valid with a sample size smaller than 4. In my view, the problem with revalidating these taxa (they recognized seven species!) is that the confidence level is way too low to make assumptions about the population sampled and their variation. It is irrelevant to talk about the PSC in this case, but it is very relevant to examine their arrangement and their analyzed material and critically conclude if we shuold follow a taxonomic assumption based on such a small N.

In general we should say that any taxonomic assumption based on small number of specimens may be ill-founded. I should add that if we have enough materials for some populations (taxa) and a few from some populations (because they are more rare or extinct) yet the few available specimens can say a lot if they are found to be divergent from the better-known taxa.

Gippoliti Spartaco you really consider that this is true: “... there is a morphologicall revision of Ungulates by Groves and Grubb (2011)”? Despite my respect for Colin, that publication to which you refer to as “a revision” is nothing but arbitrary splitting based on no tangible data at all. In that proposed classification that you somehow refer to as “a revision”, which is not, there are no data, no analyses, not even lists of specimens examined that could otherwise allow colleagues to evaluate the author’s notions… Groves and Grubb (2011) is a dogmatic monograph that could not be defended in any possible way, and its value is in the review of literature and synonyms (both of which are very valuable). To me is hard to believe that some people defend the classification in that monograph as if it were the result of scientific work—when it actually resembles religious scripts—pure dogma, no data—to the point of blindly using its classification without questioning it.

Dear Eliécer, it is not my way of thinking to blindly use anything, including any review or revision of taxonomic nature. Groves & Grubb (2011) certainly suffered the history of the project with the early death of Peter Grubb in 2006. That said, I have deal in some papers about the issue of taxonomy of large mammals in the Old World. Although I fully understand your position, I do not see many serious papers on the issue (i.e. giraffe) in the last one hundred years. Further, while giraffe species may be 4 (or 8?), certainly the trend is toward recognising allopatric differentiated populations as distinct species, in agreement with Groves & Grubb. Regarding your reference to religious script - I think the atheist Colin will smile of it. Incidentally, I have worked with Colin regarding bongo taxonomy and he had a full database with all available specimens and measurements. I hope (I'm almost sure) that such database will be made available to future researchers.