(Fwd) Re: Variable character states (Use of the rank of forma)

[Zdenek Skala]

Hubert Turner wrote:
> I feel that a distinction should be made between the actual traits
> observed in individual specimens (flowers scarlet, flowers pale red,
> flowers pinkish, flowers purplish, flowers white) and the character
> state assigned to the operational unit. The latter should be the state
> with which it was 'born', rather than a subsequent modification of that
> state. Thus, the unit might be assigned the state 'flowers red' even
> though individuals vary in their particular shade of red, and some are
> actually white-flowered due to a (recent?) mutation.
This is the crucial point, in my opinion. Two problems:
(1) which is the character state with which the species was "born"?
The "original/advanced" polarization should be an *output* of the
analysis, not the input to it, but at the same time, we need
homogeneous units... so it seems to be a circulus vitiosus here.
The frequency of the morphs within populations is not informative
enough (see BTW the classic polarization problem at the species
level where the "frequency criterion" was completely abandoned).
(2) many polymorphisms can be "relic" so to say, i.e. they are
inherited from the parental lineages. Hence, the sister groups can
share common polymorphisms instead of common homogeneous
character states. Typical example: seasonal variants in the tribe
Rhinantheae (Rhinanthus, Melampyrum, Euphrasia ...). All species
of the genus have early and late flowering morphs that are
distinguishable by several characters and are similar accross
species. Natural explanation: polymorphism is inherited
through phylogeny. So can be other polymorphisms and,
consequently, there exists no unique character state with which
the species was "born".

> Cladistics is about
> inheritable character states, which in the end boil down to particular
> DNA (or RNA) sequences. And U's or C's don't come in many different
> shades. No variability there, they actually are U's or C's.
Sure, perhaps there is some misunderstanding here. The
discussion is not about the "shades" or heritability but about the
infraspecific (between-individual) variation. There exists surely
variation at the level of DNA/RNA that is responsible for the
morphologic variation. So sequences, U's and C's have no
"shades" but still they are variable between individuals and can be
perfectly inheritable at the same time.
Best!
Zdenek
++++++++++++++++++++++
] Zdenek Skala
] e-mail:
] skala at incoma.cz