>> The problem illustrated by David's fish example is a consequence of > > our having broadened the PhyloCode's operational definition of the > > species category to accommodate those who do not accept the idea that > > a species is a population lineage segment. > > Really? Here's my example again: > > --+--A > `--+--B > `--+--C > `--+--D > `--E > > A and E can interbreed with each other, but all other combinations don't > work. > > All nodes should be considered parts of the species A + E; they were all > able to interbreed. The species is (multiply) paraphyletic, not > polyphyletic. Does that not count as a lineage segment? It counts as two lineage segments: one from the basal node to A; the other from the basal node to E. It could be considered a single lineage on an unrooted tree, but such a tree is not really a phylogeny until it's rooted (you can't identify clades on an unrooted tree). Of course, if A and E interbreed (actually, not only potentially), the two lineages are incompletely separated. On the other hand, if they are not at least partially separated, then there is no A and E in the phylogeny (only AE). Nonetheless, I would not argue that adopting the view that a species is a separately evolving population lineage segment will remove all disagreements about species boundaries and thus about the application of clade names defined using species as specifiers. > >> I am not suggesting that we revisit that decision, which I personally > > support, but if we retain a broad operational definition of "species" > > in the code, we may simply have to accept that differences in species > > concepts or hypotheses of species boundaries will result in > > occasional changes in the application of the names of low-level > > clades whose contents coincide with or overlap the membership of > > those species. As Kevin pointed out to me in our discussions of this > > problem over the past couple of weeks, variability in the application > > of clade names due to differing conceptualizations of the species > > used as specifiers is analogous to variability in the application of > > clade names due to differing phylogenetic hypotheses. > > I disagree. Disagreements about phylogenetic hypotheses are (hopefully) > scientific disagreements -- different datasets supporting different > topologies, that kind of thing. Disagreement about which species concept > should be called "species" are purely semantic. That's, as I said, > _precisely_ what the PhyloCode is supposed to stop. One of the most > important ideas behind phylogenetic nomenclature is to turn the question > "does organism X belong to taxon Y?" into a scientific hypothesis. On the contrary, disagreements about phylogenetic hypotheses are often very similar to disagreements about species circumscriptions in terms of their scientific (or unscientific) bases. Thus, just as different datasets may support different topologies, different datasets may support different inferences about species boundaries. Moreover, just as different species circumscriptions may result from people adopting different species criteria (e.g., intrinsic reproductive isolation versus fixed character state differences), different phylogenetic hypotheses may result from people adopting different optimality criteria (e.g, parsimony versus likelihood). I agree that phylogenetic nomenclature makes questions about taxon membership scientific hypotheses (rather than mere conventions); however, scientists don't always agree about the answers to such questions (conclusions), because they don't always agree about underlying assumptions (premises). > > Consequently, I still think species shouldn't be used as specifiers. > Phylogenetic nomenclature is about clades; mixing species in causes > trouble. >
> CPN mailing list > CPN at listserv.ohio.edu > http://listserv.ohio.edu/mailman/listinfo/cpn
(740) 593–9381 | Building 21, The Ridges
Ohio University | Athens OH 45701 | 740.593.1000 ADA Compliance | © 2018 Ohio University . All rights reserved.