[Taxacom] On genera
Richard.Zander at mobot.org
Tue Jul 23 16:06:05 CDT 2013
I appreciate your clarification. We are still doubtless talking past each other but less so.
You say "ancestors in a tree are assumed. They are real in as far as the daughter lineages originated from a common ancestor, but lacking said ancestor its phenotype is assumed based on the characters of the sister lineages. As such, the ancestral condition is a hypothesis based on the character distribution of the daughter lineages. "
I am referring to ancestral taxa, named taxa, sorry if I was vague. You may be referring to ancestral populations. If so, we are both right but talking past each other. If the common ancestor is the same taxon as one of the daughter lineages, then its phenotype is based on one daughter lineage, that is, itself, not the other lineage. If you take (AB) and the ancestral node is taxon A, then the node is imaginary as far as a "shared ancestral taxon" is concerned, but real if populations are concerned.
You say " Both sister taxa are equally related to their most recent ancestor." Not if one of them is the same as the most recent ancestral taxon.
You say " This statement presupposes that: a)you KNOW the ancestor in all aspects and can compare it to the descendant lineages." No, perfect knowledge is not a supposition in science. But I can infer in many cases that one of a sister group pair is the ancestral taxon by information not in the data set, e.g. chromosome numbers where there is no evidence of diploidization, ancient habitats versus recent habitats, and the like. Such scientific inferences must ace out postulation of a ad hoc ancestral taxon different from the two already at hand.
You ask " Is a longer neck more important evolutionary than a 3 aa insertion in a digestive enzyme?" Yes, it is if the molecular trait is not evolutionarily informative, such as being essential to one or a set of imaginary nodes (ancestral taxa pupping multiple daughter taxa, which often CAN be inferred from morphology and other info), or being in only one of several paraphyletic molecular lineages most of which are extinct or unsampled.
Stasis is the basis for taxonomy. Species last thousands and millions of years, and are doubtless not speciationally quiescent during that time given extant, known paraphyly.
You say " My view is that phylogenies allow you to arrange characters in such a way as to exclude certain hypothesis." Well, no they don't. There are lots of non-phylogenetically informative traits that are excluded yet evolutionarily informative. Also, a cladogram node is not a hypothesis, it is a contrivance to ensure full resolution in a cladogram.
Good exchange, thanks!
Richard H. Zander
Missouri Botanical Garden, PO Box 299, St. Louis, MO 63166-0299 USA
Web sites: http://www.mobot.org/plantscience/resbot/ and http://www.mobot.org/plantscience/bfna/bfnamenu.htm
Modern Evolutionary Systematics Web site: http://www.mobot.org/plantscience/resbot/21EvSy.htm
UPS and FedExpr - MBG, 4344 Shaw Blvd, St. Louis 63110 USA
From: taxacom-bounces at mailman.nhm.ku.edu [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of JF Mate
Sent: Tuesday, July 23, 2013 3:35 PM
Subject: Re: [Taxacom] On genera
Thanks Richard, sorry for the delay. I realised that I employed rather colloquial (and confusing) choice of words in my previous post, so I will
rephrase: ancestors in a tree are assumed. They are real in as far as the daughter lineages originated from a common ancestor, but lacking said ancestor its phenotype is assumed based on the characters of the sister lineages. As such, the ancestral condition is a hypothesis based on the character distribution of the daughter lineages. I hope this makes my position clear. In regards to the other points.
"Related, in phylogenetics, means distance on a cladogram."
Both sister taxa are equally related to their most recent ancestor. You may choose to argue that according to your metrics of choice one changed less than the other ('less evolved', different branch lengths), but this doesn´t mean it is more or less related.
"The only way that one might postulate an extinct shared ancestor is if there were data supporting such, for instance, a group of specialized taxa in isolated recent environments that share some distinctive set of traits that might be ascribed to a more generalized and widespread but now extinct ancestral taxon."
This statement presuposes that:
a)you KNOW the ancestor in all aspects and can compare it to the descendant lineages. Outside microorganisms I don't know of any case were you can compare all features of the ancestor to its descendants.
b) your metrics (aspects of the pheno/genotype you have coded) are unbiased and representative of the evolutionary rate of the whole organism. There are only a handful of organisms for which we have this information but actually coding it in a meaningful way is very subjective. Is a longer neck more important evolutionary than a 3 aa insertion in a digestive enzyme?
"Do not put your trust in anagenetic change. Always look for stasis first, because if it is there (and fits theory by biogeography, relative age of habitat, relatively generalized morphology, maybe even fossils), then postulating unknown, unnamed, and ad hoc shared taxa is not parsimonious"
I am unsure how you look for stasis a prori. Surely you are making assumptions with some sort of data (a mental tree)?
"You indicate that my "hunch" may be countered by demonstrable phylogenetic relationships. Piffle. My "scientific hypothesis" can never be countered by imaginary shared ancestors. "
I wasn´t talking specifically about ancestors or even about your specific ideas that was just another colloquialism. My view is that phylogenies allow you to arrange characters in such a way as to exclude certain hypothesis. So in this way they are a test, albeit an imperfect one as they (the cladograms) rely on bits of data to reconstruct the past so they too are fallible and subject to constant retesting.
"The stochastic element generating the shared ancestral node is provided by the false idea that characters themselves evolve (as opposed to an organism).
I would say genes evolve, albeit mostly in an indirect way as they are acted upon through the phenotype (I am not getting into the wonderful world of genomic evolution). Similarly the organism evolves indirectly. It is this indirectness, and the difference in rates, that makes speciation fuzzy. It is a gradual process, beginning with some sort of gene-flow disruption (at which point we are still talking about the same species, even if from the point of view of the genes the split has begun) and ending in distinct species (whatever they may be). Between both ends you will find the full gamut of species concepts. At the beginning a cladistic concept and at the end probably the typological SC. I am quite comfortable with a juxtaposition of states but building a phylogeny (in the most lose term of the word) involves extracting one thread and like with Schrondinger´s cat, simply by looking through your data of choice will determine where the line will fall in that fuzzy edge.
More information about the Taxacom