Re: Can we ever recover ancestors? (Was: Fwd: Are dinosaurs really reptiles? (2))

[evelyn sobielski <koreke77@yahoo.de>]

> And _then_ comes the stochastics. If we find a
> metataxon ( = lacking autapomorphies) that is older
> than its possible descendants, what is the
> probability that it really is an ancestor? The two
> considerations here are
> -- the completeness of the fossil (most potential
> autapomorphies are in DNA or soft anatomy and don't
> fossilize)
> -- and the completeness of the fossil record in time
> and space: how probable is it that we haven't
> overlooked an even more suitable candidate (or that
> it was never preserved in the first place)? This
> point is what I was talking about.

And then there is the problem of "ancestor of what?".
It can be seen in some crown taxa. For example,
_Zonotrichia atricapilla_ and _Z. leucophrys_ are
maintaining their distinctness on a
gene-pool/population level, but each individual of
either is liable to have an individual of the other
(they separated during some glaciation of the last ice
age) in its linear ancestry.

The node between these two extends for at least 10.000
generations in time and roughly 1.000 km in space, and
it apparently has quite a lot of internal structure.

Such cases are not the rule, but they are certainly
not exceptions (and in some clades, similar weirdness
*is* the rule). That is why I am just as
unenthusiastic about PN as about LN: *both* approaches
fail in just too many cases.


Eike


      Machen Sie Yahoo! zu Ihrer Startseite. Los geht's: 
http://de.yahoo.com/set