[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Species was Re: Pantydraco and the worst dinosaur name



 Microbes cluster their genomes about a wild-type, just like "proper"
 species do. They do this for many reasons, which boil down to chance,
 adaptive niche occupancy, or the ability to regularly share large
 parts of their genomes (especially plas[m!]ids) in a functional
 manner. These are, I submit (and have published) the very same
 explanations for animal species. To assert that "species" doesn't
 make sense except in metazoans and metaphytes is provincialism of the
 highest order. It is in effect to say that most life doesn't come in
 kinds, which is self-evidently false. Sure, they have their own
 modalities of being species, but that's as much an evolved trait as
 legs or spines.

Most species concepts, as far as I know, are applicable to all of life. The problems I see lie elsewhere:

-- No two species concepts yield the same results. Depending on the concept, there are from 101 to 249 endemic bird species in Mexico -- it's not like species were somehow obvious in animals with fur or feathers either! I'm sure that many, perhaps most, of these concepts measure useful parameters, and that all of them result in "kinds" that "cluster their genomes about a wild-type". They just measure parameters that are orthogonal to each other, and the "kinds" their application results in have nothing in common except the word "species". -- As a very simple example, some species concepts allow for subspecies, others make them completely impossible. Just a few days ago I saw a paper that said in passing (paraphrasing) "we advocate raising all subspecies to species rank -- we're against subspecies, because subspecies are evolutionarily uninformative". -- We were all told a lie in highschool. Few if any neontologist taxonomists, even when they work on animals with obligatorily sexual reproduction, use either of the two Biological Species Concepts: both of those are much too time-consuming to test, they describe a plesiomorphy the loss of which is usually selected against, and so on. In vertebrates at least, various phylogenetic species concepts are nowadays fashionable when DNA is available, otherwise phenetic concepts appear to be used or implied (to simplify, "if I can consistently and reliably tell them apart, they're different species"). The shift from gross phenetic to phylogenetic concepts is, in addition to genuine new discoveries of previously unknown populations, responsible for the recent large increases in species counts of tetrapods. I have a big fat book about extant vertebrates (called "animals" on the cover!) from 1984; it claims 4008 species for mammals, 8600 species for birds, 1 species of tuatara, more than 6000 species of squamates, "only about 3000 species" of amphibians, and about 20000 species of "osteichthyans". More recently, I've read of 5700 mammals species, close to 10000 bird species, 2 species of tuatara, easily 6500 squamate species, 6806 species of amphibians as of today ( http://www.amphibiaweb.org/ ), and easily 25000 species of teleosts -- again, not as estimates of how many there might be, but as approximate counts of how many have been named and are considered valid. -- All rank-based codes of nomenclature require that all organisms that are to go by an official name must be referred to a species. The zoological and botanical codes, however, do not require that authors pick any species concept -- not just not any particular one, but no species concept at all need be mentioned or implied. You can act like Linnaeus and pretend species all self-evident. There needs to be a morphological diagnosis, but that diagnosis need not be correct by any criterion, it just has to be there. (...And for extant plants, it has to be in Latin. That's right: for _extant_ plants, not for fossil ones.)

IMNSHO, the use of "species" as if it were a technical term is misleading. It leads people to, for instance, saying "speciation" when they mean "cladogenesis" (which is usually not the same under many species concepts), and even to count species as an estimate of biodiversity.

 The requirement that bacterial species should be culturable is a
 technical rather than a theoretical requirement, one that is under
 challenge as metagenomic studies appear to show that less than 10% of
 bacterial species are culturable, since they are tied into obligate
 commensuality with other species in bacterial communities. This is
 surely a flaw of assay rather than the things themselves.

Yes, but it's cemented in the Code. Never mind technical or theoretical ones, it's a _legalistic_ requirement. Until the Code will be changed, the rate at which bacterial and archaeal species are named will not noticeably increase; and the same holds for higher taxa unless the PhyloCode ends up taking over.

The PhyloCode will allow using specimens that do not belong to a named species as specifiers for clade names. In other words, it will allow us to have "genera" without species. I predict this will become very popular among vertebrate paleontologists -- most species concepts, and arguably all interesting ones, require population sizes that we simply haven't got in the vast majority of cases.

 We once couldn't identify the various distinct Leopard Frog species
 either, until we had the assay techniques to do so. They were still
 species.

Yeah, under some species concepts. Not under all of them. Different species concepts describe different, overlapping entities.