RE: EARLY EVOLUTION OF 'BIRDS'

["Stewart, Dwight" <Dwight.Stewart@VLSI.com>]

        -----Original Message-----
        From:   John V Jackson [SMTP:jjackson@interalpha.co.uk]
        Sent:   Saturday, November 07, 1998 5:46 AM
        To:     Stewart, Dwight
        Cc:     dinosaur@usc.edu
        Subject:        Re: EARLY EVOLUTION OF 'BIRDS'

        --Original Message-- From: Stewart, Dwight
<Dwight.Stewart@VLSI.com>Date: 07
        November 1998 00:30


        y.

        But how come the skew is one way for J pterosaurs and the other for
"J
        maniraptora"?

        I don't know.  Perhaps it wouldn't be.  But circumstances that are
not accounted for can prove to be very pertinent.  For example, differences
in lifestyle & circumstances of death could skew the statistical liklihood
of preservation.

        >the point is that statistics can be very
        >skewed or even rendered inaccurate by sampling.  This is
complicated by
        >interstitual or matrix inbedded chaos.  No matter how ordered a
system is,
        >there is always an element of chaos inbedded within the structure.

        There are few systems so chaotic that stats can't be applied.


        Yes, there are a few.  And they fall under the category of Chaos
Theory.
        But it's a matter of degree, really.  No known "system" (or lack of
system, really) is 100% chaotic.  In most of these cases, statistics can be
applied to a very constrained subset.  
          
        >The reverse is also true.  In a complex system, which all
biological sytems
        are,
        >there are numerous
        > Elements of chaos (or unpredictability) within the structure.
        > One element would be variation in animal behavior.  The point
being
        >that a sampling population could seen intuitively robust, but in
fact be
        >seriously skewed to produce one outcome or another.
        >
        > Dwight

        Skews are just more material for a statistician to work on.  Skew is
even a
        formally defined statistical entity.  It seems very strange to me
though
        that small pterosaurs can be preserved and found by the dozen but
20k+
        theropods either didn't preserve well or weren't so easily noticed.
There
        just possibly *might* be some very wierd process hiding those "J
mani's" but
        if you have to assume something, it's best to assume something 99%
likely
        rather than the 1% alternative.

        JJ

        What you are saying is intuitively sensible.  But there are animals
in the wild which are much larger than 20kg, for which remains are rarely
found.  A good modern example is bears.  If they die of natural causes,
their remains are rarely found.  But the remains of raccoons, which are
considerably smaller than a bear, are often found.  All things being equal,
I would say that you are probably correct, but there could well be hidden
variables (such as habits
        And lifestyle) that could result in the statistical results being
skewed.  The raw data certainly seems to mitigate in favor of your position.

        Dwight