Place in taxonomy

[pierre deleporte]

A 16:41 02/07/2002 +1000, Robert Mesibov wrote :


>Back in February my post "Why is Taxonomy Placeless?" included the para:
>
>"Spatial locations, like morphological and molecular characters, are not
>randomly distributed in the world of Life. Yet locations, unlike other
>characters, aren't part of the input in analytical methods for guessing the
>Tree's structure. Why? Geography figures large in all the evolutionary
>mechanisms (micro and macro) I've seen proposed, but it seems to vanish when
>taxonomists attempt to clarify the results of evolution. Place becomes an
>afterthought: 'This is taxon A, distinguished by these apomorphies, related
>to taxa B and C in such and such a way. Incidentally, it lives in the Atlas
>Mountains.'"
>
>Tom obviously agrees with my summary of the current situation, and now I
>have an answer of sorts to my "Why?", namely: "all phylogenetic insight does
>indeed come from the evidence of heritable characters."
>
>This is an extraordinary statement and I'm hoping Tom or another Taxacomer
>might like to expand on it.

Phylogeny is currently inferred from a data matrix of terminal taxa +
inheritable biological characters. The phylogenetic tree is an
interpretation of this data set in terms of the history of speciations
leading to diverging lineages. This historical inference necessitates not
only a model of character evolution (some notion of how evolution occurs
for characters in taxa, unlkess you cannot decide whhich tree fits the best
to your data set), but also that characters considerd to be "the same" be
effectively "the same" in the data matrix.

Using spatial locations as characters is obviously problematic. A same
spatial location may obviously be, or not be, "the same thing" historically
speaking. Kangaroos, horses and camels presently stand in australia. Their
character "present place of living" is exactly the same. But it is not the
same in terms of history, and this is indistinguishable from "directly
inherited" common location (for two kangaroos for instance).

"Standing in some place" is logically not considered a reliable character
for phylogeny inference because inherited (and possibly relevant) shared
location is indistinguishable from non-inherited (and thus misleading)
shared location. Both consist in exactly the same thing: "standing in some
place at some time". Informative and misleading locations are
indistinguishable in themselves.

Some "intrinsic" inheritable biological features may of course be
misleading (homoplasies on a tree were a priori coded as homologies), but
the assumption in this case is that most of them could be distinguished on
the basis of closer examination, while "standing in some place" is hardly
open to more detailed scrutiny.

Extrinsic or essentially epigenetic traits are thus excluded from
phylogenetic inference: they are not supposed to reliably carry the track
of common ancestry.

>  I would have thought there were legions of
>instances where phylogenetic insights were aided and abetted by knowledge of
>where the taxa came from.

Tis may be a hint, but standing at the same place is hardly a criterion in
itself. Standing at the same place AND looking like one another is more
reliable, because of the second part of the evidence, but it could lead to
the misleading overlooking of some close relatives living elsewhere (wide
range dispersal IS possible).

>  How many times have I heard plant and animal
>taxonomists describe a group as "Gondwanan," for example? Is "Gondwanan"
>simply a shorthand way to refer to heritable, placeless character states in
>that group, or is something else ringing bells in the minds of the listeners
>(or readers), like the notion that evolution in the group was constrained by
>geography?

Gondwanan may be a classification proposed after the biogeographic
analysis, not a criterion before the analysis. Unless it designates fossils
standing in so-called "gondwanan geological layers".
Also, groups may, or may not, be constrained by different geographical
features. Some dispersed, some not. The character "stands there at this
time" is no indication in itself.

I feel like saying truisms... the important point in my view is the
possibility of stating on reliable homology, that is detailed similarity
and assumed inheritance, which is hardly defendable for location data.