Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!bellcore!decvax!linus!philabs!cmcl2!lanl!unm-la!unmvax!nmtvax!shipman
From: shipman@nmtvax.UUCP (John Shipman)
Newsgroups: net.rec.birds
Subject: Re: Bird Taxonomy
Message-ID: <936@nmtvax.UUCP>
Date: Thu, 13-Feb-86 15:53:48 EST
Article-I.D.: nmtvax.936
Posted: Thu Feb 13 15:53:48 1986
Date-Received: Sun, 16-Feb-86 05:13:25 EST
Distribution: net
Organization: Zoological Data Processing
Lines: 70
Summary: This month's _Sci. Am._ describes a world turned upside down

## Jamie Andrews (...!ihnp4!alberta!ubc-vision!ubc-cs!andrews) writes:

## There's an interesting article in the latest _Sci.Am._ about the
## current wisdom on avian taxonomy.  It lists several results due to the
## method of determining what percentage of the genetic material of one
## species is shared by another.  Does anyone know whether this is now the
## standard method of taxonomic classification?

I have no degrees in biology, but my friend Leonard Compagno is a
working systematist and I have spent many an hour talking with 
him about taxonomy.  There is no guarantee this is 100% accurate
but it should answer the question; I welcome corrections.  (I'm
still waiting for someone to find the huge blatant error I posted
in a list of lumps & splits recently; I suspect no one read it.)

Leonard says there are three main schools of thought in taxonomy.
All use CHARACTERS, defined as features that differ among
individuals; for example, some popular characters among
birds are bill length, foot structure, songs, behaviors, etc.

(a) EVOLUTIONISTS try to discern the lines of descent.  They
    try to deduce whether various character states are DERIVED
    (recent) or PRIMITIVE (characteristic of the ancestors
    of the line).  The more primitive characters, the closer
    to the ancestor.  Tricky, because characters may change
    state back and forth several times.

(b) CLADISTS try to group forms by shared characters.  The
    result is a "cladogram", kind of like a tree with the
    ancestor at the root and the modern forms at the leaves.

(c) NUMERICAL TAXONOMISTS throw all the characters into a
    program that works out the degrees of relatedness.

I asked Leonard which school he prefers to make his decisions
about shark taxonomy.  He replied, "I like to try all three
and see if they agree".  In well-defined cases, they tend
to agree pretty well; if they don't agree, one should examine
more characters--but this requires diligence and hard work.

Ornithology has historically used quite superficial characters,
especially plumage; the tendency to keep most bird collections
as skins exacerbated this problem.  Internal characters such
as muscles, skeleton, and especially the musculature of the
syrinx (voice box) are important but too seldom studied.

Sibley & Alquist, the authors of the _Scientific American_
article, hit this situation like a hail of grenades in
the seventies:  they turned bird taxonomy upside down.
A lot of the groupings, especially higher taxa such as
orders and families, were set long ago and due to tradition,
politics, or inertia had never been revised.  

The work of Sibley & Alquist is largely responsible for most
of the dramatic rearrangements of the 6th Edition of the AOU
Check-List.  For example, the 6th Edition has two huge
families that absorbed what used to be eleven families:

Family Muscicapidae - includes kinglets, Elepaio, thrushes and
  Wrentit, all formerly in separate families
Family Emberizidae - includes wood warblers, Bananaquit, tanagers, 
  cardinals, true sparrows, and "icterids" (now icterines), all 
  formerly separate families.

The DNA-DNA hybridization work of Sibley & Alquist was not accepted
at first, but research that went where they pointed has tended to
confirm their conclusions.
-- 
John Shipman/Zoological Data Processing/Socorro, New Mexico
USENET: ucbvax!unmvax!nmtvax!shipman   OR  ihnp4!lanl!unmvax!...