Phylogeny Friday -- 1 February 2008

Bad taxonomy is the root of all evil. This is the correct translation of the well known biblical statement.

Once again, taxonomy is meaningless.

Spoken like someone who deals with one species. :P

Perhaps this just reflects my ignorance, but why does it matter whether or not the taxonomic names reflect the actual phylogeny?

I think debates like this come up from time to time. A few years ago it was a major issue, that human, bonobo and chimpnazee should be classified into the same genus, because the high degree of similarity between them (I believe you commented on this issue as well). This would of course have meant a)eg. chimpanzee becomes Homo troglodytes, and b)all the human-like fossils would also have needed re-classifying. But I guess in the end tradition always rules over nomenclature, especially when it comes to important species like D.melanogaster or P.troglodytes.

Perhaps this just reflects my ignorance, but why does it matter whether or not the taxonomic names reflect the actual phylogeny?

because if taxa are paraphyletic, they are useless.

that's the major difference between the Drosophila issue and the Pan/Homo issue. Pan, Homo, and Gorilla are all monophyletic taxa. Drosophila is not.

Spoken like someone who deals with one species. :P

I deal with a genus: Drosophila. I have a species with which I do most of my current work, but I do work with a group of organisms with as much genetic divergence as the entire eutherian clade. I'd hardly call that working with one species, even though I do only work with a small snippet of all life (eukaryotic or otherwise).

Perhaps, following the current code of the ICZN, which state all priority starts with Linnaeus 1749 Systema Naturae 10th edition, the fly should be named Musca cellaris?

Taxonomy matters. Your analysis is only as good as the data you put into it. Put shit species names in, get shit meaningless analysis out.

The Musca genus is pretty diverged from the Drosophila genus.

I deal with a genus: Drosophila. I have a species with which I do most of my current work, but I do work with a group of organisms with as much genetic divergence as the entire eutherian clade. I'd hardly call that working with one species, even though I do only work with a small snippet of all life (eukaryotic or otherwise).

Point taken, but this doesn't much change the fact that such a limited taxonomic scope is likely behind your statement that "taxonomy is meaningless"; I deal with the Kingdom Animalia, so taxonomic information is quite important.

that's the major difference between the Drosophila issue and the Pan/Homo issue. Pan, Homo, and Gorilla are all monophyletic taxa. Drosophila is not.

The larger issue was that chimpanzees and gorillas were in a different family (Pongidae vs. Hominidae), which was paraphyletic.

Taxonomy is in the unenviable position of serving two masters. On the one hand the goal of taxonomy is to provide us with stable, unambiguous names for organisms, at all taxonomic levels. The other goal is to order taxa to reflect their relationship. Ideally one can look at a list of taxa, arranged into various species, genera, families, etc., and reconstruct the latest hypothesis of their phyogeny. Because hypotheses of phylogeny change as we learn more, the two goals of taxonomy are at present incompatable.

As I recall, the 10the ed. of Systema Naturae was published in 1758. The second part of the species name, the so-called trivial epithet, stays with the type specimen of the species, the name bearer, forever, whether the species moves from genus to genus, or into or out of synonomy. The Law of Priority (which Darwin disagreed with) says that the first name is the right name. So, if, in fact, the type specimen of M. cellaris is the same as D. melanogaster, then the proper trivial epithet for your fruit fly is cellaris, not melanogaster. And it should be in whatever genus it should be in.

I'm of the opinion (moderately well accepted) that all the great apes belong in the family Hominidae, but that the present living genera are morphologically different enough that they need not be synonmized. There seems to me, as a spectator, that there are more fossil genera than there are.

Or we could just dump the binominal system and the Linnean hierarchy and refer to this species as melanogaster
Meigen, 1830 (or use some other uninominal species naming convention that produces an unambiguous name).

That name would be stable regardless of the phylogenetic position of the species, assuming the alpha taxonomy itself is not also a problem. If it was a member of the clades Diptera, Sophophora, or whatever, you could further clarify which melanogaster you mean by also referring to the clade (or clades -- you can use more than one) that will be meaningful to your audience -- Sophophora for fly geeks, Insecta (and maybe Diptera and Sophophora, as needed) for other people.

You know. Just sayin'.

BTW, I got the author and date from Wikipedia, so it may be bogus, but it makes the point.

Hey Frank, that is suggestion 14,781 that we do something different than what we now do in taxonomy. Just kidding, and I may be off by a few hundred. Interesting thing: there has been an official list of common names of fishes of US and Canada for many years now. It is put out, and revised every so often, by a joint committee of the American Society of Ichthyologists and Herpetologists and the American Fisheries Society. It is amusing that the common names are very stable compared to the scientific names. For example, at one time there was a large genus of minnows, Notropis, and all the Notropi had shiner as part of the common name. Now Notropis has been split into a number of genera, but they are all still shiners. The golden shiner, probably the most common bait minnow, however, was never a Notropus. Can't have everything, I suppose.

Hey Frank, that is suggestion 14,781 that we do something different than what we now do in taxonomy. Just kidding, and I may be off by a few hundred.

Certainly true! Of course, the fact that there have been lots of proposed overhauls of taxonomic practice doesn't mean we shouldn't overhaul it! ;-)

It is amusing that the common names are very stable compared to the scientific names. For example, at one time there was a large genus of minnows, Notropis, and all the Notropi had shiner as part of the common name. Now Notropis has been split into a number of genera, but they are all still shiners. The golden shiner, probably the most common bait minnow, however, was never a Notropus. Can't have everything, I suppose.

Sounds like most of those changes in the scientific names are due to splitting of Notropis into several genera, thus (under the current system) requiring most of the binomens to change. But if you just looked at uninomens, I'll bet few if any of those changed (even if a species was split into two or more species, some subset almost certainly retained the original uninomen). Scientific names can be very stable, if we eliminate the currently required association between species and generic names.

Storm the databases! Liberate the uninomens! Viva la revolucion!

I was thinking about this some today and came to the conclusion that taxonomists don't see any problem. We are generally well informed on the groups we work on and know about the name changes, etc., so we just regard them as a normal part of life. One species I described will soon be moved to its fourth genus. I know why I put it in its original genus. I know why it has moved to a second, and then a third genus (done by the same colleague). I've reviewed the MS moving it to hopefully its final genus. Looks good to me.

One problem with going to the species epithet is that different species have the same epithet: grandis, brazilensis, olivaceous,etc.

One problem with going to the species epithet is that different species have the same epithet: grandis, brazilensis, olivaceous,etc.

I hear you -- how many vulgarises are there? -- but one way to avoid that is to have the species epithet include the author and date (and page of publication, if necessary to make the name unique) -- melanogaster Meigen, 1830 is a unique name. This approach was described by Dayrat et al. (Suggestions for a new species nomenclature. Taxon 53:485-491), resurrecting and slightly modifying a proposal by Lanham from 1965 (Uninominal nomenclature. Syst. Zool. 14: 144). Dayrat et al. even discuss melanogaster Meigen, 1830 a bit.

I haven't kept up on this (or any) literature as much as I should, so I haven't heard any counterarguments, but this seems like a swell idea to me, and it's one option included in the latest draft of the PhyloCode for dealing with species names.

So we go to a trinomial: species epithet + author + date. Hopefully none of the ideas will catch on.

So we go to a trinomial: species epithet + author + date. Hopefully none of the ideas will catch on.

I still think it's better than the current system, where the species name is required to change in most instances where the species changes clades (hence the unnecessary brouhaha over Drosophila -- or is it Sophophora? -- melanogaster and Fugu -- or is it Takifugu ? -- rubripes). That has never made sense to me.

Practically, there are several ways a new system could work. At the beginning of a paper (or talk, or conversation) you could just make clear to your audience which melanogaster you're talking about -- say, the one within Diptera, which is a subclade of Hexapoda (that's one way to do it -- use more inclusive clades that people know and that are presumably more phylogenetically stable than Drosophila/Sophophora) or melanogaster Meigen, 1830 (another way).

After that, you could just use the uninomen melanogaster, if you wanted. Personally, I do this anyway -- I virtually never think or talk about Alloteuthis africana. To me, it's the only africana in the group I work on (loliginid squids), so I just think of it as africana, and I suspect that's pretty common. Anyway, at that point, it's just a communication issue. With the author/date/page information, the species epithet is decoupled from any hypothesis of relationships, and thus is unique and stable (or as stable as alpha taxonomy ever gets). In databases, you just include all the information required to make every epithet unique, and a lot of that information is already in many taxonomic databases.

Fortunately, this isn't a complex or contentious field like rocket science. It's only taxonomy -- a field filled with relaxed, friendly, mature researchers -- so I'm sure we can easily work something out that will make everyone happy.... ;-)

Make that relaxed, friendly, mature stick-in-the-muds.