"not"-lonely Dan Janzen
kinman2 at YAHOO.COM
Thu Mar 25 23:03:04 CST 2004
As for the second sentence, I just want to underscore something that I might not have made clear in previous posts about my perspective on this stuff. Simply having whole-genome sequences of all relevant species does NOT provide the answer.... Those algorithms are a long way off, as far as I can see (but still within my optimistic 3-4 decades perspective.
I largely agree with what you have said. But although I have previously expressed my own reservations about present algorithms, the huge variety of sequences in whole genomes (not to mention the order in which those sequences are arranged) will probably be informative enough to overwhelm algorithm deficiencies in groups like Diptera (where morphologists have already made great progress). For Diptera (now that molecular work has begun in earnest), I think one decade will largely resolve familial relationships and also a large proportion of generic relationships. Ribosomal RNA sequences alone could certainly be very deceptive, but whole genomes are a whole different ballgame. Thus in this case I am more optimistic.
As for reptiles, molecular work may have settled the relationships of groups like turtles (are they actually archosaurs or something more primitive as traditionally believed?) within a decade, but the overall classification and phylogeny of reptiles (which includes many major fossil groups) will indeed take many decades to unravel, even at family level. Same goes for Class Amphibia. Class Aves and Class Mammalia are more intermediate----probably substantial progress in a 10-20 year time frame. For invertebrates it really depends on how much morphological groundwork has been done (and whether the group has major fossil groups). The time-frame will vary a lot from group to group, and Diptera is one of those groups that should progress relatively rapidly.
But again, even for Diptera, I am not sure total cladification (abandoning a paraphyletic Nematocera) will necessarily be the best option. As I have demonstrated, you can encode ALL cladistic relationships and still have a few stabilizing paraphyletic groups in there as well. There will always be a certain level of uncertainty and lack of information that makes paraphyly useful (if not necessary). And that's just from the viewpoint of a taxonomist. If we had a perfectly complete record of all fossil species, a totally paraphyletic system would actually be required, but extinction and a relatively poor fossil record render that moot, so occasional (explicit) paraphyly is the optimal approach in both the short and long term.
More information about the Taxacom