Re: stratocladistics

[chris brochu <cbrochu@fmppr.fmnh.org>]

> If McCollum's analysis is correct and more broadly applicable (big "ifs" to be
> sure) 


To be honest, I am not sure I agree with McCollum.  She seems to confuse
phylogenetic independence and morphological/developmental linkage.  I
agree that morphological linkage raises red flags for the independence
of a character, but if characters arise at separate nodes, they can be
treated as phylogenetically independent regardless of whether there is a
common developmental mechanism underpinning them.

> how much worse off are we when we include stratigraphic information?


I'm not sure there's a simple answer to your question, as the influence
of stratigraphic data (as with any data partition) will depend heavily
on the relative strengths of the data partitions involved.  If the
morphological signal is very strong, the addition of another signal
(stratigraphic, molecular, whatever) will have minimal effect.


>  Stratigraphic data does suffer in that the sampling is neither uniform nor
> necessarily independent.  I'm not advocating Fox's approach which quite
> possible places too much reliance on stratigraphy.  However, I don't see why
> the consideration of at least relative stratigraphic position might not be one
> of several sorts of information to consider.  Is the quality of this data any
> worse than the information we already use for cladistics?


What do you mean by "sorts of information to consider?"  There are
several ways in which stratigraphy can be used - to help reconstruct a
tree (as Fox and others advocate), to select from a set of most
parsimonious trees (as Fox et al.s paper actually demonstrated), to
reject a most parsimonious tree in favor of a less optimal alternative
(sensu Wagner and others), to assess confidence in a strict consensus
tree, and so on.

Alternatively, one could take the view that phylogeny informs
stratigraphy - that it tells us where are sampling is very bad.  This is
the view I normally take, though I have also advocated the use of
stratigraphy as an external comparison metric once trees are obtained
from biological information.  (Bear in mind, the methods we use to get
trees from biological information assume, implicitly, that the
characters we use are heritable.  Stratigraphic rank is not heritable.) 


chris




-- 
----------------------
Christopher A. Brochu
Department of Geology
Field Museum of Natural History
Roosevelt Road at Lake Shore Drive
Chicago, IL 60605

voice: 312-665-7633  (NEW)
fax: 312-665-7641 (NEW)
electronic:  cbrochu@fmppr.fmnh.org