Another group (in this case, pair) of scientists have come up with a species concept. In this case, it's published in the Journal of Mammalogy [304kb PDF], and it turns on species being protected gene pools.
It's not new. It's not even a repeat of prior concepts. It's a collation of several species concepts, given a single name - the "Genetic Species Concept". And the authors know this: unlike many such papers they give citations as far back as the teens and 20s.
The concept they give relies on the recent explosion of techniques for sequencing DNA, however, and that is new. New technology, that is, not new philosophy. The authors, Robert J. Baker and Robert D. Bradley of the Department of Biological Sciences and the Museum, Texas Tech University, say:
...it is not our goal to establish a new concept. It is our goal, however, to organize ideas from the published literature into a pragmatic perspective to explore patterns of genetic variation in mammalian taxa and to accurately identify species and species boundaries to better understand mammalian systematics, evolution, and biodiversity. This perspective is intimately linked to the power of resolution that DNA-sequence and molecular data provide to document species boundaries, hybridization, monophyly, introgression, and so on.
An admirable goal. They define genetic species as "a group of genetically compatible interbreeding natural populations that is genetically isolated from other such groups." They contrast it with the (non-existent) Morphological Species Concept, "the" Phylogenetic Species Concept (there are many of these, not the one) and of course the so-called Biological Species Concept. The latter is better named the Reproductive Isolation Species Concept, since it is the existence of reproductive isolating mechanisms that define a species under that conception.*
I am pleased to see that the authors reject the Barcode Initiative, which attempts to identify species with a single gene, the cytochrome oxidase II gene. They rightly note that a single gene is insufficient to identify clusters of genomes.
But I have some problems with their view. Let me briefly list them.
1. This is designed to work with mammals. So far as it goes, that's fine, but will it work with other groups, like fungi, plants or corals? If it is supposed to be a resolution of the species problem, it has to apply across the board.
2. What is the "right amount" of genetic difference for something to be called a species? This is a longstanding issue (not for genes, but the same logic applies). Darwin said that there is no set amount of difference that marks all species out. Why, for example, is >5% or so sufficient in their table? This problem was also encountered with the morphological clustering of phenetics. If you chose another variable or threshold (the "phene" level), you got different answers. So which parts of the genome are to be sampled here? Why cytochrome b?
3. An organismic cluster that we call a species will have differences all across the entire genome. Why not, apart from operational reasons, compare the entire genome? Why bias the matter by looking for particular genes? After all we want to identify the natural groups here, not artifactual ones. But entire genomes are hard to cross compare. We used to do this with DNA-DNA hybridisation, seeing how much of the genomes paired up, but that is a blunt instrument. Now we do this with sequence pairings, but that involves a lot of work and judgement. Moreover, the "speciation genes" (Wu, see refs) may be different in different cases. Large amounts of gene difference may not cause speciation, while small amounts may.
4. This may break apart evolutionarily tied groups. By this I mean that while at a given point in the history of a species there may be distinct groups on the basis of haplotypes, these populations and subdivisions may coalesce over time.
I sum, while I like the methodology very much, this proposal makes a virtue of a necessity forced on us by a lack of data and insufficient instrumental capacities. But until, and unless, we have a clear idea for each group what genes are actively causing genetic isolation, it is still only a partial solution to the problem.
The authors offer something very like the Phylogenetic Concept of Cracraft here, with the different being the character set chosen as the basis for diagnosis. The arbitrary choice of data because that is the data we have is insufficient to clarify what species really are...
Cracraft, Joel (1983), "Species concepts and speciation analysis", in R. F. Johnston (ed.), Current Ornithology, New York: Plenum Press, 159-187.
---- (1987), "Species concepts and the ontology of evolution", Biology and Philosophy 2:329-346.
Wu, Chung-I (2001), "The genic view of the process of speciation", Journal of Evolutionary Biology 14:851-865.
---- (2001), "Genes and speciation", Journal of Evolutionary Biology 14 (6):889-891.
* To be clear, here, let us understand this: the concept is species. These definitions are conceptions or definitions of that concept. The plurality of species conceptions is due to the fact that not all things that are subsumed under the concept of species fit any one definition.
- Log in to post comments
Haven't read the paper, but I've seen the talk. I walked away highly unconvinced, especially considering that COX is a mt gene. I think Baker's goal was to create a quick and dirty way to determine species (for mammals) for the sake of molecular ecology and/or conservation. Be's not interested in learning how reproductive isolation arises (ie, Coyne and Orr), or studying some biological property of species or populations. It's a utilitarian goal, but it's not science nor is it all that philosophically interesting.
I am compiling a blogroll of atheists and agnostics. Do you consider yourself to be in either of these categories? And if so, would you like to be added to the blogroll?