Wednesday, May 29, 2013

Mopping up the remaining synchronous species concepts

It is time to pick up writing about species concepts again (other posts in this series can be found under the eponymous tag). I have dealt at some length with Biological Species, Genotypic Cluster Species, Morphospecies / "typological" / autapomorphic species, and the idea that species must be monophyletic. There are numerous synchronous species concepts left before I can tackle the asynchronous ones, but unfortunately I do not feel that I can treat all of them in the same depth. Some of them are simply variants of others, some apply only to very special cases, and some I simply don't really understand. I will therefore deal with all of them together in this post.

As indicated earlier, the Mate Recognition Species Concept is merely a Biological Species Concept by another name. The Reproductive Competition Species sounds like a very odd concept - organisms belong to the same species if they are in reproductive competition with each other - but basically it is also of the BSC family because it is about forming a reproductive community. Really quite odd is the Compilospecies, but it is perhaps not even a species concept but rather a term to describe the observation of extremely asymmetric gene flow between two groups of populations, so it does not matter too much for this post.

In the phenetic and cluster species family, as one might call it, it is hard to put a finger precisely on the difference between Phenospecies and Genotypic Cluster Species. In a practical analysis they would most likely result in the same species circumscriptions but the theoretical considerations of the taxonomists in question might be different: a pheneticist is not really concerned with evolution but merely aims to group organisms by overall similarity, but the proponent of the GCSC would argue that the overall similarity of specimens in this cluster and the lack of intermediates to the next cluster is evidence of an important evolutionary process, be it selection or the existence of breeding groups.

Quite close to the "selection" interpretation of the clusters is also, interestingly, the Genic Species Concept, at least in the way promoted by Wu (2001; as so often, there appear to be different ideas of what the term means). The idea is that two groups may be considered species even if they happily exchange most of their genes, which are too similar as to matter, as long as there is an island of genetic differentiation in the genome that is presumably under strong disruptive selection. Likewise, genotypic clusters may be based on a few differentiating characters even if there is wide overlap in variability for many other characters. The problem is that it is not immediately clear to me how Genic Species would be treated taxonomically unless either (1) they are, again, indistinguishable from Genotypic Cluster Species or (2) the idea is not to have species at all (but merely genes that happen to float around in this or that organism).

Agamospecies, on the other hand, is simply a fancy name for the application of either the GCSC or of "monophyletic species" to apomictic organisms. (In this case the latter would not be as completely incoherent as in sexually reproducing organisms because apomixis is a form of asexual reproduction. In this case, there is phylogenetic structure for a group of specimens to be monophyletic in!)

This leaves two species concepts that we could examine in a bit more detail, not least because they confuse me: Ecospecies and Genealogical Concordance Species (subsequently simply Concordance Species).

To cite from the previously mentioned compilation of 26 species concepts, under the Concordance Concept "population subdivisions concordantly identified by multiple independent genetic traits constitute the population units worthy of recognition as phylogenetic taxa." Maybe I misunderstand that, but the idea behind Concordance species appears to be that there would be that a species is a group of organisms containing monophyletic alleles for several independent genetic loci. That, however, seems immediately problematic. First of all, what markers are we talking about, and how many? In some cases the answer might depend strongly on what loci you have happened to pick. Okay, maybe the answer is then "as many as possible", and well, we are always limited by the quality of our data, and our results are always tentative. Okay, granted. But second, it is well known that lineage sorting often takes longer than the time between two speciation events. In a few case I know of, my interpretation would result in ridiculously lumping species circumscriptions.

So maybe my interpretation is wrong. Maybe this is not about concordant lineage sorting in gene families, maybe it is about some population genetic type approach. But then the question immediately becomes how this concept does not reduce to the Genotypic Cluster Species Concept as applied to molecular data. Again, I am confused.

Finally, the Ecological Species Concept is one that I have never understood. Species have their own minimally different ecological niche, or "adaptive zone" if you want. Presumably an advantage is that one can recognize two populations as separate species despite them still exchanging genes (i.e. them not being Biospecies) because they show different adaptations. Several problems come to mind, although some of them may simply come to my mind because I don't grasp the concept.

First, minimally different sounds nice but once again one wonders where the line is supposed to be drawn. Surely there are all possible transitions from very obviously different niches to arguably the same niche? Second, and that may really be a stupid question, but what about completely unrelated organisms that have evolved in parallel to occupy the exact same niche? Perhaps that does not happen that often with birds and insects, but let's be honest, there is a limited number of niches available for the hundreds of thousands of species (?) of land plants. What I want to say is not that any proponent of Ecospecies would consider an oak and a pine the same species because they compete for exactly the same niche space, but merely that considerations other than niche need to enter to make a meaningful species concept, and at that point one might reasonably ask why the niche part could not simply be discarded. Because third, the concept of niches is, to say the least, itself not entirely uncontroversial. There are those who argue that niches are a human rationalization and do not really exist "out there", or at least not until an organism has evolved to "fill" it, i.e. organisms may be seen to create the niches they occupy.

So far, if one were to suspect that I have great sympathies for the Biological Species Concept and the Genotypic Cluster Species Concept and relatively little love for the others, one would be correct. Next, the asynchronous species concepts, which are entirely different things altogether. After that, I will try to summarize my own position and conclude the series.


  1. Take a look here for pictures of G. zonatus and G. hoignei.

    The write up is a little confused and I need to write the blogger about it. When introduced into the specialist aquarium hobby, G. zonatus was called "sunny zonatus", and G. hoignei "shady zonatus" because of their habitat preferences. They were never taken together until 1986 when we found them together in a large area of mixed forest and savanna which had been cleared for agriculture.

    When I started studying them, I asked, are these ecophenotypes of a single species, or two separate species? I found a constant difference in number of fin rays, and established that they would not hybridize in the aquarium. Later we found them to have quite different karyotypes, and breeding behavior in the wild.

    Phylogenetic analysis of mitochodrial DNA, two separate studies, shows them to be sister species.

    There is a similar situation with Rachovia hummelincki and R. pyropunctata. But here it is a matter of occupying waters of strikingly different basic water chemistry. I tried to hybridize them, but could not keep both species alive in the same tank. I didn't know about the water chemistry issue at the time.

  2. Sorry but I am not quite sure if I get what you want to say. It sounds to me as if this exactly what is described above: the species are not recognized only be having different niches but by being unable to interbreed and morphological differences. At that moment I wonder why that should still be called Ecospecies.

  3. I was thinking maybe ecophenotypes within a species. The term ecopsecies is new to me. It might have some use in discussing sister species who occupy very similar niches in dissimilar habitats.

    I noticed I did not give you the genus name Gnatholebias.

  4. Not saying that I doubt the utility of ecological information, merely that a species concept primarily based on niche occupancy appears to be problematic.

    When I write ecospecies or biospecies, I merely mean that as shorthand for species under the Ecological Species Concept or the BSC, respectively.