When birders get into discussions of identification and the complexities of
taxonomy, as in the case of the large gulls, it seems almost inevitable that
someone will mention DNA. I think it's important to understand that DNA is
no magic solution to the kinds of problems being discussed here. The first
point to keep in mind is variability; as someone pointed out, think of the
variation among folks in any group you know, such as your high school
graduating class; the same kind of variability exists in (probably all)
natural populations, and even if you can't see it much phenotypically,
unless you're an exceptionally good and well-tutored observer, it sure is
there genotypically. In teaching biology, and writing about it, I always
cite the work of Dobzhansky and his colleagues on the variation in
inversions of one chromosome in the fruit fly Drosophila pseudoobscura. An
inversion is a change in which a segment of a chromosome is turned around;
this species carries many inversions, and their frequency varies
interestingly across the range of the species, but the important point is
that the polymorphism in inversions exists in every local population.
"Polymorphism" means harboring more than one form, or morph, in a
population; sometimes it appears phenotypically, as in the gray and red
morphs of Screech Owls, but it's always there genotypically.

Now, for a DNA analysis of the large gull complex to be meaningful, an
investigator would have to do something along the following lines. First,
collect an enormous number of blood samples from gulls of various
morphologies over a wide geographic range; if I were doing it, I would want
to collect only from breeding individuals in particular locations, and I
would certainly avoid birds that might be wandering or migrating to
heaven-only-knows where. Second, extract DNA from each sample and try to do
a fingerprint, or a series of fingerprints, of each sample. That means
cutting each DNA sample with specific enzymes (called restriction
endonucleases) that only chop at specific sequences, and then separating out
all the fragments by gel electrophoresis. Then with all my fingers and toes
crossed, I would hope that I could find some distinctive patterns. I would
hope, for instance, that samples from nominal "Vega gulls" might show one
consistent, distinctive pattern, that nominal L. a. argentatus from northern
Europe would show another, and so on. But I would be prepared to find a
real mess, with considerable variation in DNA patterns within each
population; I would consider myself lucky if I could show some consistent
geographic variation in such variability, as in the Drosophila example. As
far as I know, no one has done anything like that kind of analysis. But
even if it were done, I would want birders to recognize at least two points.
First, that analysis is not going to solve questions of defining the species
or subspecies boundaries, especially in a complicated situation like the
Larus case, where there is apparently a ring of semispecies (or allospecies)
forming a superspecies. Species and subspecies are categories that humans
like to define, and they may not be realistic categories in all cases.
Second, if someone could sneak up on some poor gull that happened to show up
in our area and get a sample for DNA analysis, I think we would be damned
lucky if the fingerprint patterns from these birds were good enough to tell
us much about the likely origin of that particular bird.

This may sound unduly pessimistic, and perhaps I'm leaning over backwards to
stress the likely difficulties; but I think people may put too much faith in
the power of DNA analysis to answer the questions that birders face.

Burt Guttman guttmanb at evergreen.edu
The Evergreen State College
Olympia, WA 98505 360-456-8447
Home: 7334 Holmes Island Road S.E., Olympia 98503


-----Original Message-----
From: Meredith and/or Dave Spencer [mailto:merdave at vib.tv]
Sent: Thursday, January 22, 2004 9:17 AM
To: tweeters at u.washington.edu
Subject: Gull DNA



My husband listens to me tell about the debate about Slaty Gull vs.
Great Black-backed Gull and wonders if any DNA work has been done to
distinguish these species. His question would include the Lesser
Black-backed Gull (now that we think there is one at Grand Coulee.) Is
there a possibility these are all subspecies of one species? Meredith <<
File: ATT956599.txt >>