In article ,
Stewart Robert Hinsley wrote:
In message , Dave Poole
writes

The oak (Quercus) is a member of the Fagacaea and related to Beech
(Fagus) and Sweet Chestnut (Castanea). On the other hand, Silver
Birch (Betula) is the principal member of the Betulaceae and most
closely related to the hornbeams (Carpinus), hazels (Corylus) and
alders (Alnus). You have to 'climb' the classification tables as far
back as the super-order (Juglandanea) before oaks and birches come
together and even then they are remote from each other. In ordinary
language, this means that any relationship they once shared is very
distant indeed.

Come the latest classification (APG) oaks and birches (and walnuts) are
in the same order - Fagales. (But the APG classification tends to larger
families and orders - for example Primulales and Theales are subsumed
into Ericales - so the two taxa may well be much the same. According to
a 1997 paper birches are closer to walnuts that to oaks.

Given the methodology that many or most of those papers use, their
results are extremely unreliable. This can be demonstrated by the
fact that two, equally respected, people often use the same methods
on two equivalent sets of data and get incompatible (and not just
different) results.

I did ask how the APG classification was produced, and I am afraid
that the answer was a lot of hand waving using the results of the
current (broken) methodology. Doubtless most of it is correct, but
doubtless a fair amount is seriously wrong.

Oh, and in case I haven't posted this before, I looked at the
seminal book by some German, and it was statistically sound (if
INCREDIBLY tedious) - the fault is in its interpreters, the ghastly
computer programs, sloppy journals and cladistic dogmatism.

[ To take one simple example - basing a dichotomy on a single gene
is incredibly error prone, for at least half a dozen biological and
statistical reasons. Yet many of those classifications and most
of those programs seem to work like that. ]


Regards,
Nick Maclaren.