Sunday, August 2, 2015

Vicariance and parsimony in biogeography, continued

On and off I have read a few more of the papers from the late 1980ies and 1990ies dealing with vicariance biogeograhy and searching for ways to apply that (then still) awesome new idea of parsimony analysis to biogeography.

Vicariance biogeography according to E.O. Wiley

Wiley's (1988) paper entitled vicariance biogeography provides a review of how he saw the development of the field at the time. In his eyes, historical biogeography was traditionally entirely 'dispersalist' simply because when the positions of continents were still considered fixed, nobody had any better idea for how species got to their current areas of distribution except long distance dispersal.

Vicariance biogeography was then, according to Wiley, produced from the convergence of three factors: (1) the discovery of plate tectonics, which provided a new explanation for disjunction distributions, (2) the advent of phylogenetic systematics, which argued that monophyletic higher level taxa are units just as natural as species*, so that the study of the distribution of higher level taxa could be justified as a sensible area of research, and (3) "the 'discovery' of Croizat by a phylogentic systematist ... [leading] to the idea that both traditional and phylogenetic concepts of the centers of origin were wrong and that all hypotheses based on a priori centers of origin should be abandoned in favor of vicariance explanations".

This third point is the one that I really don't get. 'Centres of origin' is the idea that groups were once less widely distributed than they are today and have spread out from their narrower anestral area. The obvious problem is that this idea cannot possibly be generally wrong. Vicariance only is self-defeating because it needs a widespread anestor - and how did the ancestor get to be everywhere without spreading all across the place? And if anything ever spread then it must have spread from somewere, and that was its centre of origin.

Another problem is that the ancestral areas assumed by panbiogeographers and vicariance biogeographers often appear unrealistically large, larger than those of the most widespread extant species in the relevant groups, and likely so large that the ancestor couldn't possibly have been just one biological species.

(Just as an aside, from what I read Croizat, the father of panbiogeography, really disliked Hennig's phylogenetic systematics, but of course that doesn't mean that the next generation couldn't be more conciliatory and combine ideas.)

Despite my misgivings, I find the Wiley paper considerably more sensible than the panbiogeographers' papers I have read so far. He stressed that vicariance biogeography as he understands it "does not depend on a rejection of dispersal phenomena, only on the principle that dispersal should not constitute a first-order explanation", i.e. the preferred explanation. That sounds reasonable, at least if it were qualified with "except in the case of oceanic islands".

Rooting area cladograms from PAE or CADE

I had wondered before how users of Parsimony Analysis of Endemicity and Cladistic Analysis of Distributions and Endemism would root the resulting area cladograms. Although I cannot currently find which paper I saw it in, I now read that they were supposed to use a dummy area with all zeros, i.e. no species, as the root. Seeing that shared species are the equivalent of morphological or genetic synapomorphies in this type of analysis, such an approach is logically consistent.

But does it make biogeographic sense? To me, not really. All else being equal, it just means that the most species-poor area will be considered closest to the root. But in reality, an area will often have low diversity simply because it has low carrying capacity. This is an old insight of island biogeography: the smaller the island, the less species it can carry. But that does not mean that it is should be near the root in an area cladogram that is interpreted in a historical way. By extension, the same applies to all biota, not just islands, because any disjunct biome behaves in the same way as land masses of different size.

Footnote

*) Because they are by definition what a single ancestral species turned into when it diversified.

Reference

Wiley EO, 1988. Vicariance biogeography. Annual Review of Ecology and Systematics 19: 513-542.

No comments:

Post a Comment