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

Cladistic Idea



I've had an idea about a new way of approaching cladistic analysis. I don't know
as much about the subject as I'd like to, so forgive me if this is a ridiculous
suggestion.

It seems to me that it would be interesting to approach a cladistic analysis of
fossil taxa one age at a time. As an oversimplified example, imagine an
analysis of ten fossil taxa, named "A" through "J". Imagine that
stratigraphically, they are arranged like this:

epoch #5: J
epoch #4: G,H,I
epoch #3: E,F
epoch #2: D
epoch #1: A, B, C

The standard way to do a cladistic analysis would be to take all ten taxa, score
them for all the characters, and away you go. My suggestion is to consider the
fauna of each age only in light of what has come before.

Thus, in this example, we would first do an analysis of the earliest taxa: A, B,
and C. Once we got a tree out of that analysis, we would then add taxon D to the
analysis -- but constrain the tree so that A, B, and C had to stay in the
topology that came out of the first analysis. Then we would add taxa E and F to
the analysis, but constrain the tree so that A, B, and C had to stay where they
were placed in the first analysis, and D had to stay where it was placed in the
second analysis. And so forth, adding the fauna of one age at a time,
constraining the older taxa to stay in the topology of previous analyses.

I know that time is a variable usually left out of cladistic analysis, on the
basis that the fossil record will always be incomplete. But I think that this
is an interesting way of incorporating it.

After all, evolution has no foresight, and any new evolution of taxa has to be
built on that of the past. Thinking about it another way, would a knowledge of
how life will evolve in the future help us reconstruct how it evolved in the
past? No -- each age has to be considered on its own terms, in light of what
came before it, not what came after it.

At the very least, it would be interesting to compare the results of an analysis
like this with the results of a standard analysis.

And now here's someone to explain why this is a ridiculous idea and wouldn't
work.

Thanks for reading,
-Grant Harding