----- Original Message -----
From: Victoria N. Alexander
To: Vladimir Nabokov Forum
Sent: Wednesday, September 04, 2002 8:15 AM
Subject: RE: mimicry thread turning into a rope.


VNA:  The question about mimetic forms concerns how they arise. Do they arise gradually, through natural selection? or do they arise suddenly, by chance?(and then, perhaps, get selected).

BB: Problem 1: Surely if mimetic forms arose by chance, in particular individuals, this would not explain how a particular mimetic pattern becomes established as a species-wide trait, which after all is the real explicandum. A particular mimetic pattern could not occur by CHANCE again and again in every member of the species. The problem is to explain how the pattern becomes established as a species trait.

VNA: I am not talking about the chance appearance of mimetic forms in "particular" individuals, but about the sudden appearance of large groups of mimetic forms, that look quite different from their parents. Prof Boyd may reject the concept of "epochal evolution," but I accept it. This is a concept of which, I've mentioned already, Nabokov seems to have had his own version. For a very fine summary of the argument against gradualism, see http://www.santafe.edu/projects/evca/Papers/evrevinno.html

BB: Problem 2: Patterns like eyespots and other circular or linear markings can clearly be generated by computer algorithms, but I would have thought that the mathematics of combinatorial explosion would make it unlikely that a particular COMBINATION of patterns would be coincident in otherwise unrelated species of, say, Lepidoptera, without the selective pressure of natural selection.

VNA: I have a friend mathematician Persi Diaconis whose speciality is studying coincidences. Interesting side note: He is also quite a well-known performing magician and he has taken a special interest in explaining (or debunking) the apparently psychic abilities of other magicians and mindreaders by examining the probabilities of chance patterns. As he says "our intuitive grasp of the odds is far off. We are often surprised by things that turn out to be fairly likely occurrences." The problem Prof. Boyd refers to his so well studied it has a name. It is called "the birthday problem," and the main issue is not whether or not many things are combined but the fact that the thing being represented by the "mimic" is not specified beforehand. See Persi Diaconis and Fredrick Mosteller, "Methods for Studying Coincidences," Journal of American Statistical Association 84 (1989): 854. What may appear to be freakish luck to us is an important part of how new useful forms get discovered. I think Nabokov was trying to tell us that one should not try to explain it away. Most people, I think, are irrational when it comes to thinking about luck, that is, as Diaconis argues, most people don’t seem to have an intuitive sense of probabilities (myself included: I have a special fondness for coincidences). It is trivial to say that any one unspecified member of the 12,000 or so species of butterflies will by chance resemble another unspecified species or an unspecified object in nature. It would also be trivial to predict that someone's lottery ticket (a combination of numbers) will match the winning number in a given drawing without saying who will win. Although there is nothing magical about winning the lottery, every winner cannot help but feel a bit favored by fortune. This is because, as far as the lottery winner is concerned, he or she did prespecify the winning number.

BB: Problem 3: If similarities of the order we find in mimicry could be explained largely by spontaneous pattern-formation, would we not expect there to be similarities, as marked and as frequent as those of accepted mimicry, linking random features of random species, species that are both phylogenetically and geographically unrelated? Does this in fact occur with anything like the same frequency as mimicry...?

VNA: Yes. Contemporary evolutionary biologists have found, for example, that the fact that many animals have four limbs instead of, say, five or three is not indicative of common ancestry. Four limbs represent a common configuration among mammals and birds because the laws of pattern formation have the tendency to produce tetrapods. As Brian Goodwin writes,
"Tetrapod limbs are defined as the set of possible forms generated by the rules of focal condensation, branching bifurcation, and segmentation in the morphogenetic field of the limb bud. ... The idea of a common ancestral form as a special structure occupying a unique branch point on the tree of life ceases to have taxonomic significance. Now tetrapod limbs could have arisen many times independently." See How the Leopard Changed its Spots: The Evolution of Complexity (New York: Scribner, 1994), 155.

BB: Problem 4: If predator-prey relations play an insignificant part compared with spontaneous pattern formation, would we not expect to find that there are cases of elaborate visual mimicry where predators cannot see the pattern involved? For instance, in night-flying moths would there not be proportionately as many cases of elaborate mimicry of wing-patterns in the undersides of the wings (the upper sides of the wings of course are often camouflaged to escape detection while the moths rest by day) as there are on the wing patterns of day-flying butterflies, even though in butterflies predators can be duped by the result and in the case of night-flying moths predators like bats pay no attention to visual detail? 

VNA: Camouflage should not be considered mimicry. The concept of natural selection works very well to explain why so many moths have camouflage coloring. Moth wings are patterned, naturally having dots and lines. They are more regular than tree bark. The wing patterns that are less noticeable against tree bark are those that are most busy. A simple bold pattern would stand out. Busy wing patterns and bark have relatively the same degree of "information," as they say in theoretical physics. The salient point is that camouflage wing patterns do not resemble bark. As H. Frederik Nijhout points out, "there is no requirement that the elements of the pattern be of a specific shape." See The Development and Evolution of Butterfly Wing Patterns (Washington: Smithsonian Institution Press, 1991), 236. Because any busy wing pattern of various configurations of dots and lines may be selected as camouflage, natural selection might very easily and gradually evolve a large number of various kinds of complicated wing patterns that would function as camouflage. If natural selection had only one pattern that it could select for fitness, for example a pattern that looked like the wing pattern of another insect, then the chance that natural selection would find it would be relatively low.