I am not sufficiently up to date with the current state of the
taxonomy wars to offer other than a very general response to Andy
Anderson's remarks on Richard's pippit. Parenthetically, though,
I have been a birder of sorts since at least 1981 and I am still
amazed by the capacity of this type of debate to kindle the kinds
of controversy that in the old days would have led to gloves
across the face and pistols at dawn. It's a bit silly, really.
Instead let me respond with a short pertinent book review of
something I read over the break. I'll include a short, very
general remark on the Richard's pippit issue below.
Review: The Beak of the Finch - Evolution in Real Time, by
Jonathan Weiner, Vintage (Random House), 1994.
The book is an alternating, quasi-narrative account of the work
of Peter and Rosemary Grant, their family and students, with
snippets of related work by others and with a kind of selective
potted history of the theory of evolution itself.
The Grants have worked intensively on Darwin's finches on a small
island in the Galapagos since 1972. The island is large enough
to contain a variety of niches, but small enough to allow them,
with sufficient dedication, to tag, measure and record the
physiognomy and breeding history of virtually all finches of the
two main species on the island. The species are sedentary, so
there is no significant emigration/immigration effect. The work
is fascinating and their conclusions are even more so. They
conclude, among other things:
1. Changes in morphology can be fast and dramatic, from a
statistical perspective at least.
2. Miniscule statistical changes in beak size can lead to
measurable advantages or disadvantages in a population sense,
3. Hybrid matings between closely related finch species (?) are
not rare, and may act as a source of genetic variation for
natural selection to use.
And so on. Their period on the island - they spend six months in
Princeton and six months on the island each year, roughly - has
seen some very dry years and very wet "El Nin\~o" years, which
push evolutionary changes in the finches fairly rapidly, but in
opposite directions, leading to a kind of stuccato equilibrium.
Although the work of the Grants is very impressive and the
conclusions very interesting I found the book itself rather
disappointing in many ways. It avoided coming to the point far
too often for my liking. This may be a writer's way of
increasing the entertainment value of the book, to keep the
reader guessing what was coming next, but I was more interested
in the science than in the drama, such as it was, and this device
simply irritated me.
The science, when it did come, was sanitised of all mathematics
and statistics - even though the conclusions were fundamentically
about statistical effects - and the reader is spoon-fed small
snippets almost surreptitiously, as if the author expected the
reader to find scientific information somehow unpalatable and
mathematical and statistical ideas repellent. Computers, on the
other hand, are wonderful devices that in themselves can solve
all problems. He even holds up Darwin's well known mathematical
weakness as a kind of virtue. He seems to regard Darwin as being
to Biology what Newton was to Physics, and who would disagree,
but goes on to say that Darwin's theories are much more
accessible than Newtons because they were expressed in "everyday
language that anyone could read, whereas Newton's ideas were
expressed in Mathematics", as if Mathematics were some kind of
The word "statistics" is assiduously avoided, although the entire
story is quintessentially statistical. I suspect the author
would be horrified to learn that one of the scientists he
casually includes in the discussion on first name terms - Steve
Arnold - is in fact a statistician with no biological training
(just like me, in fact).
If you are not put off by all this pussy-footing around the issue
I'd say read the book, because the material is both interesting
and important, even though it comes frustratingly in dribs and drabs.
To return to the Ricards pippit issue, one message of the book
for me is that species are far more fluid and less well defined
than classical taxonomy would suggest. In a sense it would be
surprising that even on the Australian continent itself, with its
wide variety of niches and climatic conditions, Anthus
novaeseelandae did not have regional populations that showed
morphological and genetic variations large enough to approach the
distinct species status, however you define `species', and that
is no longer clear to me.
In a sense I find Andy's contention rather uncontroversial.
William Venables, Department of Statistics, Tel.: +61 8 303 3026
The University of Adelaide, Fax.: +61 8 303 3696
South AUSTRALIA. 5005. Email: