[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

re: PAUP vs. MacClade

As expected, almost all of the time MacClade and PAUP match, but for
some, still unknown reason, on this occasion MacClade beat PAUP by two
steps. The setup is the simplest possible with no tinkering with the
settings. Always going for the shortest tree. Takes seconds usually.

Jaime wrote:

Note the Second: The output is only as good as the input. If one is
assessing the whole of Diapsida or Sauropsida, one had better have more
than 500 characters, since there are a LOT of animals being assessed
It is also wise to avoid suprageneric clades, but to use species as
specifiers, and list all which specimens were coded from in an appendix
table, for the sake of reference. This is my one major gripe with
recent (and thought provoking) analyses.

DP replies:

While one could take thousands to millions of characters, (including
width of proximal second metatarsal vs. midshaft diameter of tibia, if
present) the present work is finding a single tree with about 150
characters and 116 taxa, mostly basal forms and enigmas. The cool thing
about PAUP is it will find a tree with any three taxa, be they lungfish,
gold fish and cows or what have you. PAUP can be fooled. It will find a
relationship come hellorhighwater. So, retracing with MacClade and
looking at hypothetical sister taxa side by side is a great help and a
great test.

I've been fooled many times, as I've discovered recently as I make
corrections. But, the thrill is in the learning -- and I am learning
about burrowing snakes, turtle kin and ichthyosaurs -- things I admired
but never studied until now. Hence the earlier questions about snake
diphyly (sp?) which is becoming more and more interesting evrery day. I
can see where the arguments have their strengths, but again, I think the
arguments can be solved by expanding the taxon list.

On Jaime's second point I could not agree more. The use of suprageneric
taxa cloaks the transitional taxa between major clades, and of course,
one chooses characters from within the cherry basket of a suprageneric
taxa with complete subjectivety. Too be avoided.

One thing I have noticed is a vast emptiness in the literature when it
comes to extant lizard skeletons.

I'm also of the opinion that many modern lizard taxa are, like
Sphenodon, living fossils, virtually unchanged since the Mesozoic.

David Peters
St. Louis