Darwin (was: Phylogenetic evidence)

[pierre deleporte]

A 09:12 26/07/2002 +0200, Zdenek Skala wrote:
>From: pierre deleporte
> >But they [clades] must share characters in common, unless they would not be
> >identified as clades.
>
>Well, only to complete the argument:
>character matrix (+ = apomorphy)
>   1 2 3 4 5
>A - + + + +
>B + + + + +
>C + + + - -
>D + - - - -
>=> cladogram (parsimonious now, I hope!)
>(((AB)C)D)
>...where no character state is shared by all its members.

OK. No "face value character state" is shared by all the members. But an
equivalent feature is shared: it is the "1" state in D,C,B,and A. In A, it
takes the form of a derived "0", which is a plain historical avatar of "1".

The class DCBA is supported by the shared equivalent trait "1" = "derived
0" in A.

>...but more interestingly: Pierre:
> >Clades are classes if you consider reversals as apomorphies, what cladists
> >do as a rule. They classify organisms AFTER phylogenetic analysis, thus
> >'reversed' absence of characters, as well as "reversed" states similar to
> >plesiomorphic ones, are identified as apomorphies ...
> >Classes based on these nested sets of synapomorphies, interpreted in the
> >light of the cladogram, including possible apomorphic absence - reversals,
> >are thus classes of equivalence, not similarity clusters.
> >In my view, you can't reason for phylogenetic systematics by considering
> >"face value" character states in the data matrix.
>Pierre, do you mean that the "reversed" character state is re-interpreted
>as a completely new character state (different from the plesiomorphy)?

Yes of course, it's a logical consequence of accepting the cladogram as a
basis for phylogenetic systematics.

>  In that case you also cannot have homoplasies but only a series of
> independent (though indistinguishable) apomorphies.

In no way. Independent (though a priori indistinguishable) synapomorphies
ARE homoplasies by definition. So not only you can have homoplasies, but
you have them. If (and only if) you accept the cladogram as a reliable
phylogeny. Homoplasy is a pure product of phylogenetic analysis.

>  Consequently, all the indexes of homoplasy are useless

In no way. They give you the amount of discordance in the data, hence the
amount of "errors of identification" of characters you have to accept (or
sawllow...) in order to accpet the cladogram as a reliable phylogeny. They
participate in the evaluation of the reliability of a cladogram (also
Bremer index of support, which may have some link with homoplasy, etc...).

>  - you will have always fully resolved cladogram.

In no way. Cladograms are not always fully resolved, and not always show
low homoplasy ratios when they are resolved.

>  Is this what you want to say?

Clearly not. It's just your imagination. No offense   :-)

Best,
Pierre