The nature of cladistics [was: Whooper; Canadian geese split; paraphyly]

[pierre deleporte]

>At 10:16 AM 11/18/2004, Curtis Clark wrote:
>>My other post makes it clearer, but basically the idea is this.
>>Phylogenies are built of species. As one approaches the species level,
>>the tools of phylogenetic reconstruction become less and less useful,
>>until at some point "you're not in phylogeny any more".
>
>And Thomas Lammers wrote : Well, that's pretty much my original point! 
>Cladistic methods at those
>levels seem inappropriate BECAUSE they are divorced from the reality they
>purport to reflect.  When someone can construct an algorithm to 
>accommodate that, it will be a major breakthrough.

I think I can agree with Thomas Lammers on some general ground:
"all relevant information should be used for phylogeny inference" (at any 
level 'above' overwhelming cross-breeding in sexual organisms, and at any 
level in strictly clonal organisms... why do phylogeneticians always seem 
to overlook these nice clones, which should logically be the more appealing 
critters...).
The more we will know about evolutionary processes of all kinds, the more 
we will be able to reconstruct the evolutionary past (including phylogeny 
inference). Populational processes don't stop with the splitting off two 
"baby species", and in the case of big sized populations split into equal 
parts, we could even say that the populational processes go unchanged in 
the first times. There is just an opportunity for beginning some diverging 
genetic drift, separating the biological profiles of the two baby species. 
Maybe this is what Curtis Clark hints when he says that processes differ at 
the supra-specific level. Well, populatioonal processes both go on 
basically unchanged, and something new happens (no more panmixy, 
opportunity for diverging drift). The resulting profile of the two species 
is the product of continued processes in two recently separated biological 
systems.

This seems plain logic, but fairly general. Maybe too general to be useful...
Could Thomas specify some aspects of populational processes which could be 
of some use for phylogeny inference?
Thomas hinted 'natural selection': which way? Do we know of general laws, 
or rules, of 'natural selection' which could guide phylogeny inference? Or 
is selection too "contextual" to be of general phylogenetic interest?
Also "reproductive isolation' was suggested: same question, what is 
relevant for phylogeny inference?

My view is rather that we could make use of a better knowledge of rules (or 
laws) of 'mutation', if any, e.g. substitution processes in genomics. But 
isn't this already done presently, some way? (e.g. parsimony weighting 
schemes, or ML models, for transitions versus transversions to say the 
least...). Do you suggest that more populational studies could help 
improving our appreciation of such substitution processes?

Just inquiring...
Pierre








Pierre Deleporte
CNRS UMR 6552 - Station Biologique de Paimpont
F-35380 Paimpont   FRANCE
T?l?phone : 02 99 61 81 66
T?l?copie : 02 99 61 81 88