Although Vladimir Nabokov may be better known for his outstanding literary achievements, particularly as the author of the novel Lolita (1955), he had an equally impressive genius for science. While acting as curator at Harvard's Museum of Comparative Zoology in the 1940s, he became an expert on a group of butterflies popularly known as "Blues." He named one species and several have been named after him. He published nine articles on lepidoptery in a number of prestigious scientific journals. During this time, he also developed compelling opinions about evolution. He argued, rather heretically, that some instances of insect mimicry did not result from Darwinian survival strategies; that is, slight resemblances could not be furthered by the function or purpose they served, leading gradually to better resemblances. I contend Nabokov was partially correct in his belief. Recent advances in evolutionary biology, namely structural evolution and neutral evolution, can be shown to support his argument. I also argue it was Nabokov's aesthetic interest in the mechanisms behind teleological phenomena that gave him the insight he needed to construct a theory of mimicry that was quite progressive for his time.
In Nabokov's novel Pale Fire, the protagonist, John Shade, suffers a heart failure. When he recovers, he recalls having seen a fountain while clinically dead. Although he had been a skeptic prior to this experience, he begins to suspect that there is an afterlife. Later, he becomes even more convinced when he reads in the paper of a Mrs. Z who also saw a fountain while on the other side. He feels that this is too unlikely of a coincidence not to be meaningful. Shade travels to meet her and learns there had been a misprint in the article. She had seen a mountain, not a fountain. But ultimately Shade is not disappointed. He declares,
It dawned on me that this Was the real point, the contrapuntal theme; Just this: not text, but texture; not the dream But topsy-turvical coincidence, Not flimsy nonsense, but a web of sense
Some thing or someone seemed to be making "plexed artistry" or "ornaments/of accidents and possibilities." Apparently, whether or not there truly is a God or an afterlife is not as interesting to Nabokov as the fact that it is suggestive coincidences that give the impression life is like a novel with an omniscient and somewhat playful author.
As an artist, Nabokov found teleology a natural mode of analysis. Teleology is, after all, originally derived from an analogy with the concept of artistic intentionality. The term telos has been variously interpreted as end, purpose, function, utility, motive, intention, goal, or design. It must be noted that Nabokov understood telos differently from how it is often understood today, that is, as a handmaiden to divine Providence or as a predetermined "linear" mechanism. There are many different versions of teleology. Nabokov tended to be sympathetic to the teleologies of, say, 20th century vitalist Henri Bergson or the 19th century Kantian teleomechanists. Nabokov recognized that teleology, like art, involves two distinct aspects, mechanisms for maintaining order and mechanisms for discovering new order, what I call directionality and originality, respectively. Systems in nature are formed according to mechanistic laws that arise spontaneously from quantum indeterminacy. Yet these law-abiding systems also come to function in advantageous ways not predicted by those laws. These two aspects, emergent lawfulness and ability to transcend laws, makes natural systems seem telic, that is, progressive or creatively organized toward goals. This view of telic behavior can also be used to describe artistic behavior. To some degree, art must be a spontaneous engineering-like activity that synthesizes already existing elements according to known laws. This would correspond to the aspect of directionality. However, if a work of art were completely directional it would be too predictable. Art must also involve essentially unpredictable activity that while conforming to mechanistic laws can transcend them. This might be done through misinterpretation or by making use of a chance analogy. In this view, when activity involves both directionality and originality it is intentional, which, in this analysis of Nabokov's work, is used synonymously with artistic and telic.
The mechanisms behind insect mimicry resonated deeply with Nabokov's interests in art and teleology. To put it simply, mimicry was for him a concrete exemplar both of art and of telos, as he understood the concepts. It was "the chance that mimics choice, the flaw that looks like a flower." One cannot, he argued, rely upon Darwin's idea of functionality and selection to explain the origins of the coincidental resemblances involved in mimicry.
Darwinism explains how a certain structure has evolved by pointing to the advantage it gives or the purpose it serves. Although Darwin did away with the Divine Watchmaker, as Richard Dawkins has noted, natural selection takes on the role of Creator as the Blind Watchmaker. In this sense, Darwinism actually encourages a kind of teleology as a mode of biological explanation. Nature may not fulfill God's design, but it still fulfills a design shaped by function, which ultimately derives from increased rates of reproduction.
Nabokov tried to encourage lepidopterists to consider, on the one hand, the "nonutilitarian" mechanisms of physical and chemical constraints involved in wing pattern formation. These mechanisms can help explain, for example, the similarity between viceroy and monarch butterflies, a phenomenon once known as Batesian mimicry. Nabokov claimed that neither species benefited from the resemblance, so it seemed a case of design without purpose.
On the other hand, Nabokov also tried to encourage lepidopterists to consider other "nonutilitarian" mechanisms behind the creation of a butterfly that looks like a dead leaf (fig. 1). As far as he was concerned, this form of apparent mimicry is something of a "statistically insane" coincidence, not a useful representation by one individual of another or of its environment. In this case, he argued the resemblance could not be attributed to a mechanistic cause (laws of pattern formation, as in the viceroy-monarch relation), any more than it could to an efficient cause (duped would-be predators, as in natural selection), but must be attributed to a coincidental cause.
According to Nabokov, both viceroy-monarch mimicry and dead-leaf mimicry involve elements of chance. Noting this is essential if one is to view these phenomena as telic in the sense that Nabokov understood teleology. In what follows I demonstrate that, despite the fundamentally stochastic processes involved in the evolution of separate species (such as the viceroy and monarch), non-inherited family resemblances do emerge that indicate the existence of, what, in the literature of teleology, is called a ground plan. I further argue that the creation of dead-leaf mimics is attributed to selectively neutral mechanisms that happen to produce very unexpected coincidental similarities between two genetically unrelated organisms, an insect and a plant.
Nabokov had a profound respect for coincidences as coincidences. One of his favorite examples of a selectively neutral instance of "mimicry" was a butterfly wing marking that looked like a drop of dew with a glint of light reflected in it. As he described it, a line along the wing edge running through the "dewdrop" was shifted in a perfect imitation of refraction – masterfully rendered, but still a coincidence. It is difficult to imagine what function or advantage could be ascribed to an imitation of a dewdrop on, say, a Blue's wing. It must be admitted, then, that some forms of "mimicry" may be imposed by the lepidopterist's powers of interpretation. Since such cases of false mimicry conferred no reproductive advantage – it merely amused – Nabokov notes that it "seemed to have been invented by some waggish artists precisely for the intelligent eyes of man."
Although Nabokov was careful not to align himself with any particular philosopher, the ideas reflected here are Kantian. In Kant's Critique of Judgment, he argues that patterns in nature are typically appreciated in one of two ways, with the aesthetic judgment or with the teleological judgment. The former can appreciate nonutilitarian forms of spontaneous organization – beautiful geometric patterns on butterfly wings, for example. The latter can appreciate utilitarian forms that exist because of the function they serve within a system – drab coloring on moth wings that makes them virtually invisible to predators, for example. Like Kant, Nabokov also felt that such aesthetic or teleological phenomena only seemed to be indicative of a divine creator. Kant insisted that no empirical evidence could ever prove the existence of a transcendent being: both aesthetic and teleological judgments were valid for the reflective not the determinate judgment. However, whereas Kant's ultimate goal was to make some kind of argument for the existence of a divine creator, Nabokov seemed content just to relish the feeling that nature sometimes seems purposefully created. As Nabokov writes in his now infamous passage on mimicry:
The mysteries of mimicry had a special attraction for me. Its phenomena showed an artistic perfection usually associated with man-wrought things. Consider the imitation of oozing poison by bubblelike macules on a wing (complete with pseudo-refraction) or by glossy yellow knobs on a chrysalis ("Don't eat me – I have already been squashed, sampled and rejected"). Consider the tricks of an acrobatic caterpillar (of the Lobster Moth) which in infancy looks like bird's dung, but after moulting develops scrabbly hymenopteroid appendages and baroque characteristics, allowing the extraordinary fellow to play two parts at once ... that of a writhing larva and that of a big ant seemingly harrowing it. When a certain moth resembles a certain wasp in shape and color, it also walks and moves its antennae in a waspish, unmothlike manner. When a butterfly has to look like a leaf, not only are all the details of a leaf beautifully rendered but markings mimicking grub-bored holes are generously thrown in. "Natural selection," in the Darwinian sense, could not explain the miraculous coincidence of imitative aspect and imitative behavior, nor could one appeal to the theory of the "the struggle for life" when a protective device was carried to a point of mimetic subtlety, exuberance, and luxury far in excess of a predator's power of appreciation. I discovered in nature the non-utilitarian delights that I sought in art. Both were a form of magic, both were a game of intricate enchantment and deception.
More than a few commentators – for example Brian Boyd and Stephen Jay Gould – have supposed that Nabokov's rejection of natural selection as an explanation for mimicry was motivated by a belief in creationism or a belief in a predetermined telic principle of organization. Lepidopterist Charles Remington claims that Nabokov's lack of mathematical training prevented him from realizing how quickly a singular reproductively fit mutation can spread throughout a population. However, Nabokov had no quarrel with the basic principles of natural selection, generally speaking. He simply thought they did not apply to mimicry. He insisted that, in addition to reproductive fitness, there were other mechanisms that drive evolution.
Due to the fact that the subtleties of teleological arguments are these days not well understood, Nabokov's biographers have tactfully avoided what they must find to be his rather embarrassing teleological leanings. In Nabokov's Blues (1999), Kurt Johnson and Steve Coates describe the thoroughness and reliability of Nabokov's research at Harvard. They give a detailed description of Nabokov's controversial method of classification, which was based more on internal anatomical structure than external appearance; importantly, they do not note the significance of his focus on the symmetrical nature of certain structures, a quality that can have little direct bearing on reproductive fitness but relates instead to teleology and its search for the laws of biological form.  The recent Nabokov's Butterflies, edited by Brian Boyd and Robert Pyle, collects Nabokov's many scattered references to lepidoptery and includes a significant previously unpublished piece on teleology and evolution entitled "Father's Butterflies"; however, in their commentary, the editors do not attempt to unpack Nabokov's theory of mimicry. This is what I will attempt to do here.
First of all, for any commentator to assume that Nabokov's argument against gradualism reflected a belief in creationism is to jump to the wrong conclusion, since there are several well-known neutral (or "nonutilitarian," in Nabokov's terms) evolutionary mechanisms that assist natural selection. These include the mechanisms that involve random drift, various "laws" of biological form, and spontaneous pattern formation. Secular forms of teleology focused on these kinds of phenomena and did not seek to prove the existence of a creator, but rather sought to understand the principles that govern biological form and activity. Judging from Nabokov's scientific work, one may conclude that he was convinced that most of nature's patterns are primarily spontaneous and produced by largely deterministic processes; most were shaped somewhat by natural selection; and a rare few patterns (specifically, resemblances) were purely coincidental.
Natural selection works best as a theory if the definition of a successful form remains open. The problem involved in appealing to natural selection to explain mimicry is that the successful form is defined beforehand. Success is not found in whatever works but in what works because it looks like something else already in existence and engaged in its own unique evolutionary journey. Thus, a would-be mimic species must strive toward a predetermined, and nevertheless always fluctuating, goal in the course of its random mutations. Nabokov recognized that mimicry in this view was flatly goal-directed, i.e., teleological in a sense with which he did not agree. (He, like Bergson, thought telic achievements were never predefined.) Furthermore, goal-directedness contradicts the very essence of natural selection as a theory, the strength of which lies in its tautological nature: whatever survives survives. Moths, whose color blends with tree bark, are relatively straightforward examples of natural selection favoring a nonspecific pattern. Camouflage moth wing patterns do not have to match bark precisely: they only have to have an appropriate degree of drabness and complication. Mimicry, however, would be a different phenomenon altogether. It, like art, would be a more particular representation of one thing by another, and it would also require audiences or observers continually to make the "correct" interpretations of the representation.
If a resemblance does not appear to have any use, like the dewdrop marking on a Blue's wing described above, it is often either written off to chance or attributed to an intentional being with a sense of humor. This kind of situation is so often played out in Nabokov's fiction it is clear that examples of "nonutilitarian" insect mimicry helped Nabokov define his concept of art. Nabokov's aesthetic theory can be summarized as the answer to the question, What makes an object a work of art, as opposed to an object produced merely by spontaneous physical causes? or an object produced by random chance? It is the suggestion of intentionality. A similar argument was used by teleologists to define telic systems in nature. What makes an object telic, as opposed to an object produced merely by spontaneous physical causes? or an object produced by random chance? Again, it is the suggestion of intentionality.
The insights Nabokov gleaned from teleology demonstrate that the value of a theory may not lie in the specific answers it provides but in the way it identifies a particular problem. Teleologists from Aristotle, to Kant, to Bergson have noted that nature seems to develop in a limited direction toward more complex and organized forms than would be possible if left entirely to the flip of a fair coin. (Unlike the causally separate events involved in a hundred tosses of a perpetually fair coin, events in nature are affected by what went before.) They also noted that nature often creates original systems that seem to anticipate unpredictable future needs by reinterpreting old tools for new uses. Therefore, teleological phenomena were considered analogous to, though categorically different from, products of human intention. They concluded that some rational guiding principle must emerge from the order and arrangement of the components of organic systems.
Despite our current discomfort with teleological explanations, teleology has made significant contributions to scientific progress. As historian of science Timothy Lenoir explains, many 19th century Kantian teleomechanists were not mystics in any sense. They simply sought to investigate how in biology a
functional whole gets assembled and why it is organized in one way rather than another ... [They believed] the whole determines the organization of the parts, [but] in so doing it never violates physical laws. On the contrary, the very existence of the [whole system] depends on the most efficient organization of the parts. The end ... determines organization ... by establishing the parameters of possible physical solutions.
The teleomechanists worked in embryology, rational and functional morphology, and cell theory. If one traces the history of what is called nonmental teleology through Aristotle and the Kantian teleomechanists, one sees that it eventually led to an investigation of the laws of biological form and pattern formation, anticipating the work of 20th century scientists, such as D'Arcy Wentworth Thompson, Alan M. Turing, and Brian Goodwin, as well as lepidopterist H. Frederik Nijhout, who have offered alternatives to an exclusively adaptationist evolutionary program.
True to his interests in nonmental teleology, Nabokov focused much of his scientific work on morphology and the ways in which butterfly wing patterns are shaped not by their fitness vis-à-vis contingencies in the external environment but by internal chemical and mechanical constraints. Nabokov sought to understand a pattern as a whole in order to determine how the organization and activities of individual parts are related to the whole. Such internal processes lend themselves to rational description – unlike the etiological myths of some Darwinian explanations. Nabokov objected to Batesian mimicry because it attempted to explain the similarity between the monarch and the viceroy in terms of its utility rather than in terms of morphogenetic constraints.
True to his interests in mental teleology, Nabokov focused much of his literary work on the role of coincidence in the evolution of meaning. For example, an external observer (a bird) can use its own internal mechanistic rules to (mis)interpret one object (a butterfly) as another (a leaf) by means of a false analogy or coincidental resemblance. In this way, an observer redefines the object, causing it to have an effect (or ascribing to it an effect) that cannot be attributed to the object itself. If the effect of a resemblance can be said to have some use to the object, it may seem caused by its utility, which leads to a somewhat different mode of teleological explanation than the one described just above. Here we have two separate systems interacting, an observer and an object observed. Here effects are not internally constrained by a natural relationship between the parts and the whole. Rules are transcended rather than followed. Mentalism relates to mystical forms of teleology that posited a divine artist who brings about coincidences for the sake of the functions that they serve.
Although Nabokov had a critical understanding of a number of different kinds of teleology, his own idea of telos, or "agent X" as he called it, seems most influenced by Henri Bergson. Evolution, in Bergson's view, is creative, not fixedly mechanistic. As he writes in 1907, if teleology implied
that things and beings merely realize a programme previously arranged .... As in the mechanistic hypothesis, here again it [would be] supposed that all is given. Finalism thus understood is only inverted mechanism.
Bergson argued that functions, dependent as they are on interpretation (or perception, to use his term) can have an unpredictable effect on the direction of evolution. In this view, telos includes the aspect of originality and must be emergent, not prespecified. Telic behavior in this view is not predictable.
Cosmic teleology concerns the suggestion of intention in the natural world. I use the term narrative teleology to designate the suggestion of intention in fictional worlds. Nabokov wrote of both cosmic and narrative telos as emergent phenomena. We can find examples of emergent telos or intentionality in Nabokov's The Real Life of Sebastian Knight. Knight's biographer is his own brother, who is simply called "V." While V is conducting his research, his habitual, and often eccentric, modes of perception sometimes happen to resonate usefully and poetically with situations that do not demand them. He recognizes that the use of such stochastic resonances, as it were, is a way of creating a new order, or an original work of art. He decides, therefore, to abandon the traditional mode of biographical research and to employ this artistic mode instead. V continues to gather information in a haphazard fashion, accumulating an excessive amount of irrelevant detail, but he remembers and pays more attention to those details that happen to form some aesthetically interesting coincidental patterns. For example, while V unsuccessfully searches for the last woman to break Sebastian's heart, he stumbles upon a woman who had given Sebastian his first heartbreak. V uses the coincidence to organize his narrative. He explains,
A more systematic mind than mine would have placed [the first heartbreak] at the beginning of the book, but my quest had developed its own magic and logic and though I sometimes cannot help believing that it had actually grown into a dream, that quest, using the pattern of reality for the weaving of its own fancies, I am forced to recognise that I was being led right, and that in striving to render Sebastian's life I must now follow the same rhythmical interlacements.
There seems to have been a law of some strange harmony in the placing of a meeting relating to Sebastian's first adolescent romance in such close proximity to the echoes of his last dark love.
When other interpretable situations arise that cause coincidences to make a kind of literary sense, V notes that the resultant complex structure seems to have been purposefully made available (from a retrospective point of view, that is). What V ends up with is a biography that beautifully and uncannily seems to mimic the life of Sebastian Knight.
The "real" and the actual lives of Sebastian represent two separately evolved narratives, which nevertheless develop similar themes. One might compare this example of, what in evolutionary biology is called, convergence with that of a hummingbird and a hummingbird moth (see fig. 2). The bird and the insect traveled separate evolutionary pathways; nevertheless, each has been shaped by the same niche. Thus, they resemble each other. To my knowledge, no one has ever made the argument that the resemblance itself confers a reproductive advantage. The hummingbird moth is not a mimic of the hummingbird. Neither the moth nor the bird required the other as a model on which to base its appearance. Similarly, in The Real Life of Sebastian Knight, V does not make much use of the facts of Sebastian's life to model the biography. Rather, he focuses on his own pathway, his own recollections and feelings.
In the end V feels the biography is a more interesting achievement than the actual life because it does somehow ring true, and such convergence is so unlikely as to seem, perhaps, supernaturally contrived. The reader realizes the irony, however, knowing that two individuals toiling in a similar niche will inevitably encounter similar circumstances, which may very well result in a "law of some strange harmony."
To summarize, this paper has so far been concerned with three types of resemblances in nature, none of which can be properly called mimicry. The first is the nonutilitarian viceroy-monarch resemblance, which, as I demonstrate below, can be attributed to their both being formed according to similar morphogenetic, chemical, and energetic constraints. The second is the resemblance between a dead-leaf and a particular butterfly, which, as I also demonstrate below, was probably created by coincidence, not the gradual refining powers of natural selection. The third, which I only mention in passing, is the resemblance between hummingbird and hummingbird moth that is created by natural selection, but which does not serve a purpose. In all three cases, the resemblance is incidental, i.e., it does not exist because of the function it serves. Resemblance then in all three examples is something of coincidence.
Although Nabokov was charmed by coincidences, there is no reason to believe he did not have an appreciation of the laws of probability. It is rather straightforward to understand that samples drawn at random may not be without some coincidental order. In his fiction, Nabokov often notes that people commonly think they have found evidence of the supernatural in apparently improbable events, good or bad luck, and even funny coincidences. As Michael Wood has argued, Nabokov never assigned symbolic meanings to coincidences, though his characters frequently did. For example, in "The Vane Sisters," Sibyl Vane and her friends tend to assume that random systems should show no regularity at all; therefore, if any kind of pattern is detected, they believe it could not have arisen by chance, and they assume that patterns without any other cause must have a patterner.
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 member of a different species or some unspecified object in nature. Similarly, it would also be trivial to predict that someone's lottery ticket 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. As noted above, what seems odd about mimicry is that the advantageous form toward which natural selection must work seems prespecified. We know that natural selection does not work this way. Therefore, if one wishes to explain the cause of a resemblance, one might assume one of two things. First, all numbers are not equally probable, that is, the nature's lottery is biased. (I argue this is the case in the viceroy-monarch relation.) Second, the resemblance only seems to be prespecified, as it does to the individual lottery winner who feels especially favored by fortune. (I argue this is the case with the dead-leaf butterfly.) The focus of teleology has been to investigate the validity of such assumptions.
Although natural selection might stabilize a resemblance once it is found, selection alone could not create it. This, I argue, is the crux of Nabokov's dispute with the Darwinists of his day. Nabokov frequently noted that a fortuitous resemblance might confer a reproductive advantage through predator evasion. The question he posed was: How do resemblances arise? gradually, by natural selection? or somewhat suddenly, by chance?
If one wanted to argue that the viceroy-monarch relation arose "by chance," one would have to discover limiting mechanisms that would make this resemblance probable. Today the goal of structural evolutionary theorists, like that of the 19th century Kantian teleomechanists, is to elucidate the "principles of organization" that result in the appearance of similar patterns in nature. They study the energetic, mechanical, morphogenetic constraints that limit the range of possibility in biological forms. Like the teleologists, structural evolutionary theorists contend that these constraints result in a relatively small number of structural archetypes considering the multi-dimensional space in which they evolve. Thus, if there were a film version of Earth's evolution that could be rewound and run again, many of the forms we know today would reappear. The task of the biologist today, then, is to discover which forms are likely to appear. Only then is it worth attempting to estimate which of them will have a differential fitness significant enough to confer a reproductive advantage.
Structural archetypes occur throughout nature. They are sometimes referred to as structural attractors and compared to Platonic solids because they exist, as concepts, prior to the process of natural selection. One should be cautious when making such comparisons, however. The term "attractor" may incorrectly imply a pre-existing physical form that draws natural processes toward it.
Nabokov argued that all butterfly patterns are "variations on a theme," which was a favorite phrase of teleomechanists. To lepidopterists in the 1940s, this usually meant a variation on the nymphalid ground plan (fig. 3), which was conceived independently in the 1920s by both B. N. Schwanwitsch and F. Süffert. The ground plan was thought to be a Platonic ideal from which all possible wing patterns were derived through the distortion of individual elements. The existence of such a template would mean a wing is not a blank sheet on which any design can be developed if it happens to be better as camouflage. Moreover, it would also mean that one need not posit a Darwinian primitive ancestor from which species diverged in order to explain similarities between families and species. The ground plan fits the Kantian idea of a teleological principle as a heuristic device, and it can be used to make predictions about evolution because some patterns are more likely to occur than others, regardless of utility.
Nabokov did not regard the ground plan as a primitive form whose individual elements were variously distorted in later descendants. He regarded the ground plan as the initial conditions and set of constraints that limit the otherwise stochastic process of pigment diffusion across a wing surface.
Nijhout has recently explained the origins of the ground plan in terms of the reaction-diffusion model developed by H. Meinhardt in 1982. Nijhout applied Meinhardt's model to a domain that resembles an individual spoke-like section of a butterfly wing (see fig. 3). The proper term for this spoke-like section is a wing cell. Nijhout's model of pattern formation within a wing cell starts with a rectangular field with the bottom side open (see fig. 4), where that part of the wing cell would attach to the insect's body.
The model assumes the existence of two substances, an activator and an inhibitor, distributed equally throughout a rectangular shape. The activator is so-named because its by-products (i.e., its syntheses) tend to increase their own production. The inhibitor is so-named because it neutralizes the activator's by-product. Thus, only if the inhibitor is present in the right amount will a homogenous steady state throughout the rectangular field be maintained.
Nijhout found that if activator is slightly increased along three of the edges, where the wing veins are, a reaction-diffusion process occurs that gives rise to dynamically stable patterns. (The steady-state behavior might go out of balance along the edges because these areas do not get the same balancing feedback from all neighboring regions as other areas do. In this way, a sudden increase in activation might occur spontaneously.) Soon after increased activator appears at the edges, it diffuses toward the center. A line of activator forms in the center then retreats toward the open edge. In the concentrated areas at the end of the receding line, activation increases even further. Then these areas are finally pinched off by surrounding areas of inhibitor, leaving traces of activator production behind. These traces, together with any traces that may have been left on the edges, form areas of greater or lesser density. This difference in density is metaphorically referred to as the "topography" of the wing cell, which in turn affects the diffusion of pigment. Individual pattern elements of the ground plan (see "major themes" in fig. 5) are determined according to whether or not the topography "attracts" or "repels" pigment, condensing or stretching it into various shapes.
Both Nijhout and Nabokov describe overall pattern formation in terms of individual wing cells. They both realized the significance of the fact that each wing cell is separated by veins, preventing communication (diffusion) between wing cells and providing a frame that shapes the reaction-diffusion process. This was not generally acknowledged in Nabokov's day. He noted that sometimes one wave of pigment in a given wing cell might happen to reach a threshold at about the same time as the waves in its neighbors, giving the impression of a band running across several wing cells. See, for example, the marginal bands and the symmetry bands in the ground plan (fig. 3). However, Nabokov argued, the bands are merely "pseudo lines," "manmade" lines, or macules in "linear disguise." He was quite adamant about this because he recognized the relevance in terms of understanding pattern formation in butterfly wings. In "The Nearctic Forms of Lycaeides H.b. (Lycænidæ, Lepidoptera)," Nabokov writes,
What we see as a transverse, more or less sinuous, "line" or "row" of spots seems to me to be the outcome of two unrelated phylogenetic phenomena. The "upper" part of the "row" ... is formed by spots having radiated fanwise ... owing to an apicoid extension of the wing texture; the "lower" part ... [has] been pulled out ... presumably by a cabitoid extension ... This is why the classical conception of a row of ocelli [eyecells] as the result of a statically placed line or band having broken up into spots seems to me absolutely irrelevant to the understanding of the Lycænidæ pattern. Insofar as spots have been evolved in this family, they occupy different positions in different species or genera, and what we see is [sic] not the remnants of a definite band in a definite place, but this or that stage of a more or less coordinated longitudinal movement of spots ... (certain comet-tail traces of this progress are sometimes caught and fixed aberrationally). In a word it is not a row of squares on a chessboard, but a shifting line of attacking pawns.
In "Notes on the Morphology of the Genus Lycaeides," Nabokov further explains how spots and "pseudo lines" emerge when progressively shorter waves of pigment reach a limit and a gradual deepening and concentration of the pigment occurs, forming a spot. Then the whole process is repeated again in a second wave down the length of the wing cell. If the process happens to be activated earlier than normal in development (by, say, some extreme temperature), an additional spot may occur. Pattern, Nabokov concludes, is the "result of those processes and not a 'primitive' line which Mother Nature automatically traced with her brush." Although Nabokov's obsession with the dynamical nature of patterns may have seemed eccentric to other lepidopterists in the 1940s, it is now clear that Nabokov was beginning to sketch out a theory of spontaneous pattern formation that was not fully articulated until the 1950s when Alan Turing published "The Chemical Basis for Morphogenesis." One might compare Nijhout's understanding of the effect of wing cell topography on pigment diffusion to Nabokov's observation that pigment seemed to be "pulled out" during development by wing "texture," diffusing from high density to low density areas. In significant ways, Nijhout's research recalls Nabokov's. Both describe pattern formation as that which, in Nijhout's words, is "not specified by a detailed genetic program or blueprint, but [as] emergent properties of relatively simple processes occurring in particular physical or chemical context." Nabokov's Bergsonian notion of emergent teleology would have made him sensitive to the existence of spontaneously formed patterns which seemed to but do not require a predetermined program.
According to Nijhout, a rather simple model "can generate virtually the entire diversity of patterns found in nature." (One must simply allow, first, that the relative values of the two diffusion coefficients or of the decay constants in the lateral inhibition model can be varied and, second, that the activator along the wing cell edges will naturally diminish as it flows toward the open end.) Presumably, the range of spontaneously generated patterns includes "eyespots." Indeed the prevalence of eye-like macules is expected given that there commonly is found a single point trace in the center of the wing cell (see fig. 5). Pigment will tend to diffuse from the point, leaving a relatively lighter center behind.
As Peter Godfrey-Smith has noted in the context of teleological arguments, "properties due to constraint" should not be considered functional because whatever accidental functionality they may come to have does not explain why they exist. One must be careful to distinguish between "fortuitous benefit and genuine adaptation." It should be possible to design an experiment to see whether or not eyespots are any more common in nature than would be predicted by Nijhout's model without the help of fitness selection. Another experiment might be designed to test how likely it would be that the ground plan constraints might spontaneously generate two separate species that exhibit a kind of family resemblance, such as the viceroy and the monarch.
While Nabokov was working at the Museum of Comparative Zoology in the 1940s, it was believed that the viceroy and the monarch had traveled two different evolutionary pathways, the viceroy guided by natural selection alone. The two species were said to exemplify convergence. Yet, unlike the convergence involving hummingbird moth and the hummingbird for example, the viceroy was said to have been shaped by the function of its resemblance to the monarch, not their similar environmental niches. According to Batesian mimicry, the fact that the viceroy looks like the unpalatable monarch makes it less likely to be preyed on by birds that have sampled monarchs. Therefore, a resemblance might be reproductively advantageous to viceroy butterflies as they would be preferentially selected. Nabokov tested both the viceroy and monarch himself and reported that they both were unpalatable. Credulous Darwinists continued to believe the Batesian mimicry story without testing it themselves. Finally, forty years later, a study by Jane Van Zandt and Lincoln Bowers (using mockingbirds to taste test viceroys) found that indeed the viceroy is also "bitter." Consequently, the Batesian theory of mimicry has been discredited, and the viceroy-monarch relation is now considered to be an example of Müllerian mimicry. According to this theory, different species of butterflies, each unpalatable, mutually reinforce the association between appearance and bitter taste.
But there is more to this mimicry mystery. A 1984 study by J. R. G. Turner  provides some support for Nabokov's argument that natural selection may not have gradually and painstakingly shaped resemblances between different species of butterflies, such as the viceroy and monarch. Turner concluded that Müllerian mimics and their models have not traveled long and unique pathways. Turner shows that because butterflies share a common toolbox (e.g., laws guiding reaction-diffusion processes) for forming patterns, a single mutation leads to a large change in appearance, bringing one species reasonably close to another. Turner's findings are consistent with Nabokov suspicions. Nabokov supposed that the resemblance between the viceroy and the monarch was the product of similar mechanistic, temporal, or chemical constraints. Nabokov never denied that functionality might help stabilize the resemblance between a "mimic" and its "model" once it was already in existence, but the initial cause of the resemblance must be sought in some ahistorical limiting principles, such as the ground plan constraints, that have nothing to do with survival or increased reproduction.
Nabokov argued against Darwin's idea of gradual adaptation toward increasing reproductive fitness, which requires each new mutation to be functional if it is to survive and replicate. It is far from clear how the viceroy species might gradually be driven nearer to a likeness with the monarch species while the monarch population is randomly mutating in some other direction. In "Father's Butterflies" (written in 1939), Nabokov ridiculed the logic behind Batesian mimicry. Selection according to function could only result in Batesian mimicry if the mimic were
pursuing this goal consciously, having conferred beforehand with the model and determined that the latter, during the full number of centuries required by the toiler at evolution toward a gradual attainment of resemblance, would remain unchanged (in the kind immobility that a painter demands of his model). The process would accelerate further if the model just as consciously indulged the imitator by mutating part way in proportion to the mime's mutations, or if the very goal of the imitator were to change concomitantly with the evolutionary metamorphoses of the model, in the same way a painter, having begun a nude of a young female model, might strive for a likeness with such ardor that, as he tirelessly recorded every trait, he would, in the end, find that he was depicting the old woman into which the model had evolved during her plurennial pose. Yet the concept of evolution in no way presupposes either the existence of a conscious and focused will within a developing creature, or a coordination of actions between two creatures or between a creature and its environment. As for the presumption that nature mesmerizes subjects selected for mimetic study, influencing them to perform specific roles, that notion must be relegated to fantasy, for where are the anchor points for the cobweb of hypnosis? The same variations that might result from a blind struggle for survival, no matter how credible their results may appear ... endlessly retard the putative course of a given evolutionary process, for it is here that the element of happenstance reappears.
Since Motoo Kimura first introduced the idea of neutral evolution in the 1960s,  research in evolutionary dynamics has recognized a number of nonadaptive forms of evolution, which can occur in a variety of evolutionary settings, one of which is a fluctuating environment. Nabokov's description of the difficulties that a bug-artist would have in representing a changeable model refers to such an environment. Darwinian natural selection cannot maintain consistent pressure in a particular evolutionary direction if the contexts in which natural selection makes its interpretation of fitness are constantly changing. Today these conditions are referred to as a percolating landscape where fitness peaks and valleys shift under a developing population's feet.
Even in a percolating landscape, a species can remain stable. In 1940s it was widely believed in biology that, without Darwinian selection to favor some forms over others, even the most prevalent structures would have no ultimate power against the tendency to disorder, enshrined, for example, in the second law of thermodynamics. Nabokov realized that this was not the case. In fact, the existence of fluctuating environments allowed him to see more clearly the evidence of other mechanisms maintaining order in the face of constant random mutations.
By considering the effects of spontaneous organization of reaction-diffusion processes and the limits illustrated by the ground plan, one can better understand the likelihood that the viceroy and monarch wing patterns might arise without the help of external agents or environmental conditions consistently favoring one form over others.
Above I have referred to the viceroy-monarch relation as a case of apparent design in nature without purpose. Thus, in Kantian terms this phenomenon is valid for the aesthetic judgment but not the teleological judgment. This type of resemblance (it can hardly be referred to as mimicry) is attributed to the mechanisms of spontaneous pattern formation. In the next section, I will consider another type of resemblance, the dead-leaf "mimic," that appears to be valid for the teleological judgment: the resemblance's apparent function would seem to be the cause of its existence. However, as I will show, the butterfly's resemblance to a leaf may only occasionally serve an accidental functionality, which does not explain why the resemblance came to exist.
Now I will illustrate how a coincidentally useful form such as a dead-leaf butterfly might have emerged not gradually via fitness selection but suddenly with the aid of neutral evolutionary mechanisms working with the ground plan. This illustration will also parallel the fact that the fossil record indicates that evolution often proceeds in sudden jumps, not by Darwinian-style gradual adaptation.
Karl Ernst von Baer, one of the most prestigious teleomechanists and father of modern embryology, may have been one of the first biologists to take notice of the effects of what is now known as neutral evolution. He was concerned with mechanisms for stasis and the maintenance of biological order. He discovered that the early stages of embryological development are both very conservative and restrictive of later development.  Current research shows that because the developmental process is integrated (as von Baer argued), mutations in the genotype will not necessarily result in phenotypic difference.
Kimura's theory of neutral evolution now expresses these ideas much more clearly.  When different genotypes result in a single phenotype, natural selection has no way of favoring one phenotypically equivalent genotype over another. Under such conditions, random mutations in the genotype accumulate unchecked for long periods of time. This results in a great amount of diversity in the gene pool, all of which is selectively neutral.
To illustrate the concept of many-to-one genotype to phenotype mappings more concretely, let's say there is a species of butterfly whose genotype for wing pattern formation can be designated as sm***s. The letters and asterisks illustrate (in a very simplified manner) a genetic sequence. The asterisks are wildcard genes. Only the first two and the last letters have to be set in a particular way (s, m, and s, respectively) for normal sm***s forms to be produced. Therefore, the four butterflies with the following distinct genotypes are virtually identical in appearance:
Despite the genetic differences in these four genotypes, they all result in the same wing pattern phenotype. The final s represents a regulator gene than can inhibit or excite the activity of the wildcard genes in such a way that the pattern produced is the same in every case. By definition, Darwinian natural selection cannot see or differentiate wildcard variability. This results in long periods of species stasis, which are followed by sudden appearances of new species.
It turns out that several 19th century teleologists were also aware of the phenomenon of punctuated evolution. While von Baer investigated mechanisms underlying genotype diversity and phenotype stasis (herein considered directionality) as described above, another teleologist Etienne Geoffroy Saint-Hilaire investigated phenotypic change and sudden speciation (herein considered originality). Geoffroy Saint-Hilaire studied teratology, i.e., aberrations in morphological development that led to phenotypically unique individuals. Based on this work, he developed a theory of evolution suggesting that morphological change was not slow or gradual but dramatic, occurring when distortions in the universal ground plan were suffered by the developing embryo. Current research now shows that large errors in the early embryonic stages set the stage for further development. The effect of the error may be replicated causing serious aberrations in form often leading to spontaneous abortion but sometimes leading to what Geoffroy Saint-Hilaire considered "monsters" and what we call mutants today.
Given his interest in teleology, Nabokov was probably familiar with this research. He, like Geoffroy Saint-Hilaire, also conceived of speciation as a sudden event that he compared to the bursting of a bubble. Moreover, the last great effort made by teleomechanists against gradualism in Darwinian evolutionary theory took place in Nabokov's home of St. Petersburg just decades before Nabokov's birth. Nabokov scholars will be interested to know that von Baer, who had moved to St. Petersburg and was active there until 1876, wrote about Nova Zembla, a remote part of Russia that figures very largely in Nabokov's novel Pale Fire. Von Baer's two articles are entitled "Sketch of animal life in Nova Zembla" and "On the recent Russian expeditions to Novaia Zemlia."
In the 1960s and 1970s, Nabokov was more absorbed with his literary career than with butterflies. He missed the opportunity to compare Kimura's newly articulated theory of neutrality to von Baer's early writings. Nevertheless, we can assume that he would have been able to do so. Nabokov's literary works involve the kind of conceptual framework needed to understand evolutionary neutrality and sudden speciation. Just as a population can accumulate a large amount of irrelevant genetic diversity (wildcards), a writer might accumulate a large amount of irrelevant and diverse detail awaiting some turn of events that might make them relevant. As "V" discovered in The Real Life of Sebastian Knight, the more diversity one accumulates, the greater the chance that something useful might be found.
In biological realm, the question to be considered then is, How does this hidden genetic diversity ever get expressed, allowing selection to "use" it? One possible answer is that hidden genetic diversity is expressed when there is a mutation in a regulator gene, which suddenly removes the limits on wildcard activity. But one might have to wait a long time for enough of the right random mutations to occur in a population before one would see a new species arise. Another possible answer involves environmental conditions.
It so happens that butterfly wing pattern development is particularly sensitive to climatic conditions. Many species have wildly different summer and winter or wet- and dry- season forms. In fact, the dry-season form of Precis almana looks a bit like dead-leaf mimic although the wet-season form does not (fig. 6). Studies have shown that near-lethal high or low temperatures temporarily disrupt the activity of certain genes during development.  (This might have been the kind of thing that Geoffroy Saint-Hilaire was investigating.) When this happens to an entire population, the previously hidden wildcard diversity can be suddenly revealed, and, in turn, natural selection can act on it.
There is, therefore, a relationship between how individual butterflies within a species can differ depending on environmental conditions during development and how new species evolve and differentiate from each other. It is worth noting here that many 19th century teleologists investigated a possible relationship between the developmental process (ontogeny) and the evolutionary process (phylogeny).
The experimental disruption of specific genes during development is now widely used to study how new butterfly species arise because it simulates the effects of natural mutation. According to Nijhout, "temperature shock could thus mimic an alteration [a mutation] in a mechanism that regulates gene expression during pattern development, and the [aberrations] that develop would reveal the consequences of small quantitative differences in the cellular mechanisms of gene regulation." Whereas mutational events are typically unique to individuals, wide-spread disruption caused by environmental conditions (extreme temperatures, for example) might affect large numbers in a population. In such a situation, if regulator genes happened to be knocked out, natural selection would have the opportunity to act on a generous sample of previously hidden genetic diversity, thereby increasing the likelihood of the aberrational forms becoming favored and established.
To illustrate more concretely how wildcard variability can be expressed, we return to our thought experiment that represented genetic sequences with words. Now let's say our population of butterflies with the genotype sm***s is exposed to near-lethal high temperatures during development. The trauma temporarily knocks out the final s gene during development, as indicated in italics, producing aberrations,
In these forms, the final s, a regulator gene, does not function. Hidden genomic variability in the wildcards is now revealed because their activity is no longer limited by the regulator gene.
Generally speaking, in most temperature-shock studies, a number of distinct aberrations appear, though on the whole, most tend to be duller in color and have smaller eyespots than the parent. The double bands of the central symmetry system (see fig. 3) tend to merge and shift toward the axis of symmetry. Marginal and submarginal pattern elements (see fig. 3) tend to be lost. In sum, compared to their parents temperature shock aberrations are more drab in color and their markings contrast less with wing background.
The remarkable result is that some aberrations closely resemble other species. These imposters are called "phenocopies."  Phenocopies look like other species, but they still have the genotype of their parents. In our thought experiment, a phenocopy is still an sm***s butterfly even though it does not express the original wing pattern. Moreover, it still produces normal offspring, like its parent.
Taking our thought experiment one step further, let's say that a temperature-shock damaged smiles happens to look very much like a dead leaf, even though its parents look nothing like leaves whatsoever. It is not as unlikely as it may seem that offspring could differ so much from their parents. As Nijhout has noted, each individual pattern element in the ground plan tends to be controlled by one or very few genes. Moreover, in 1927 Süffert deduced that only two modifications of the ground plan account for the main characteristics of the dead-leaf pattern: the distal half of the central symmetry system is aligned with the outer band of the ocelli border, forming a single line down the center of the wing (like a midline leaf vein), and most other pattern elements are muted (fig. 7). Therefore, one might easily conclude that the first dead-leaf forms, perhaps even the Kallima (figs. 1 and 7), could have arisen as phenocopies produced by trauma that knocked out one or two genes. Perhaps it is not incidental that Kallimas tend to occur in hot climates.
It is important to recall, however, that the first dead-leaf phenocopy, if such an animal ever existed, would still have been a member of its original species. It would have still had the genotype sm***s. It would have merely looked different from its parents. Despite the distortions suffered during its own development, any heat-shock aberration will still produce normal offspring like its parent. However, it has been shown that if only one particular kind of aberration is selected and interbred for many generations, eventually offspring begin to produce the aberrant forms without being exposed to temperature shock during development. This process is known as genetic assimilation.
To Goldschmidt working in the 1950s, it seemed as if Lamarckian evolutionary theory involving the inheritance of acquired characteristics might apply here, but more recent studies have shown that this is not the case.  Genetic assimilation can be explained if one demonstrates how the probability of a given mutation is increased. For example, if functionally distinct smiles phenocopies tended to segregate themselves,  the probability that part of the population might successfully mutate into toiler is higher than it would be in a population with more genetic diversity. In a population of, say, interbreeding smokes, smells, smiths, and smiles butterflies, only one forth of the genes are properly set, as indicated in bold face font, to create the toiler species. In a pure smiles population, half the genes are already properly set, as indicated in bold face font, to produce the toiler species.
How many mutational events would it take for the genetic sequence smiles to get to toiler? And what kinds of constraint does nature put on the mutation process? To investigate these questions, I will make use of a favorite game of Nabokov's, which he called "word golf." In this game, one starts with an English word (for example, smiles) and changes one letter at a time until one arrives at a target word (for example, toiler). Each letter change along the way must result in an English word.
The game happens be remarkably suited to illustrating how evolutionary change works: just as word golf does not allow the use of nonsense letter combinations, nature aborts deleterious forms.  For example, imiles is prevented from getting into the evolutionary game. Desirable mutations (smiler, for instance, in generation one, and soiler, in generation two) only have to compete against other viable forms (smiles, smiley, and stiles, for instance), not all possible forms. One should also note that there are fewer possible viable forms in a gentoypically pure population than in the mixed population. In the mixed population, the first generation of viable mutations might yield: stokes, spokes, smoker, smoked, smalls, shells, stiles, and smiler. In a pure smiles population, there are only three possible viable mutations in the first generation, stiles, smiley, and smiler. The possible viable mutations in succeeding generations would be
In a population of smiles, there is a thirty-three percent chance that a viable mutation will result in the desired direction, smiler. In the next generation within the smiler group, there is a thirty-three percent chance that a mutation will occur in the desired direction, soiler. Within the third generation soiler group, there is a twenty-five percent chance that a mutation will occur in the desired direction, this time to toiler. Thus, true toiler species might be produced in just three generations. This will happen with the probability of 1/36. It is also interesting to note that some viable mutations that go in an undesired direction may have descendents that return to an earlier form, thereby opening up the possibility of additional desired mutation.
Assuming that during this mutation process, near-fatal temperatures continue to cause trauma during development, the genotypes smiles, smiler, and soiler might be phenocopies of toiler. A toiler, however, would not require exposure to near-lethal temperatures in order to have its dead-leaf appearance. This thought experiment presents a picture of the way in which genetic assimilation could work. Although the game of word golf as a model for evolutionary processes is very simple, it does illustrate how the random mutational process might be internally constrained, as well as historically contingent.
Although aberrant forms might be stabilized through selection, neutral mechanisms allow the mutation process to diffuse through a wide range of genotypes with various wildcard configurations. Therefore, neutral mechanisms have the power to create diverse genotypes. This is the key point that I would like to make in relation to Nabokov's insistence that the dead-leaf "mimic" was not produced gradually by natural selection. It might have appeared suddenly, in a developmental response to heat shock, for example, and only later, if ever, realized an advantage by fooling predators. Neutral evolution does not deny that Darwinian evolutionary mechanisms exist. It supplements Darwinism by explaining the mechanisms for genotype diversity that increase the probability that a structure may be found that is coincidentally close enough to a form with potentially higher (or at least different) fitness.
Since dead-leaf mimicry cannot be explained in terms of a kind of family resemblance as the viceroy-monarch relation can, it is simply coincidental or "false" according to Nabokov,
i.e. attained by essentially different means. Such false resemblances are extremely rare and the number of characters involved is small, and this is as it should be, since such 'convergence' depends upon the mathematics of chance. False dissimilarities also occur (and are also rare), i.e. the striking difference between one type and another is seen, when analysed, to be due to a simple and brief process of evolution in an unusual direction.
The more we study Nabokov's lepidoptery writings, the clearer it becomes that his arguments against Darwin were not conservative but innovative. In fact, Nabokov was not against Darwinism properly understood. He simply had something more radical to add.
In his fiction, Nabokov satirizes the tendency to see utility as a cause of coincidental resemblance. For example, in "The Vane Sisters," a man is obsessed with typographical errors. He pours through the pages of books in search of misprints such as "Hilter" for "Hither." If such typos happen to alter the sentence in a way that make sense, they might seem to confer a new, perhaps more clever, function; it would seem as if some one had intended the new meaning. As noted above, Nabokov claimed such typos were illustrative of "the chance that mimics choice, the flaw that looks like a flower." Nabokov calls attention to the fact that we have a very strong tendency to perceive useful coincidences as intentional, whether the agent we posit is an artist, a supernatural being, or a duped mockingbird.
Incidentally, two of the Kallima dead-leaf's main plant hosts are Girardinia and Strobilanthes. The first plant has a leaf with a spiked edge, the second a large purple and green variegated leaf. The dead-leaf butterfly looks nothing like either of these two plants, and an egg-laying dead-leaf butterfly, decked out in her fantastic disguise, would stand out like a sore thumb. Though this might not have any bearing on whether or not a predator might still mistake the dead-leaf for a different kind of leaf fallen among Girardinia or Strobilanthes leaves, this mismatch represents one of nature's Nabokovian ironies.
Among Nabokov's readers it is well-known that he intended to write a book-length study of insect mimicry that would offer an addendum or correction to the theories of evolution that were current in the 1940s. Although he never attempted that book, hints of it appear in Ada, or Adore (1969), "Father's Butterflies" (2000), The Gift (1963), Pale Fire (1962), and Speak, Memory (1966). Nabokov's observations on insect mimicry and camouflage are especially applicable to aesthetics because they depend upon the notion of representation. Nabokov's knowledge of aesthetics also contributed to his theory of evolution. As a writer, he knew that coincidental patterns often form unintentionally in narratives. He also knew that they could appear intentional if they could be interpreted as functional. He supposed that a similar situation could occur in nature.
When Nabokov writes that he "discovered in nature the non-utilitarian delights that [he] sought in art," he was referring to beauty created accidentally and spontaneously, not to beauty created by a supernatural artist with a predetermined plan. Although some commentators have supposed Nabokov's argument against Darwinism indicated a belief in creationism, this is inconsistent with Nabokov's aesthetics, as well as the fact that he frequently expressed a lack of interest in religion. Moreover, Nabokov's choice of the term "agent X" for a teleological force indicates a departure from the then-popular notion of teleology, namely that of the 19th century philosopher Herbert Spencer, whose pernicious interpretations of Darwin tried to justify what have come to be known as social Darwinism. According to that argument, natural selection becomes a substitution for a supernatural purposeful creator, and evolution progresses inevitably toward more perfect forms, superior races eliminating inferior ones in accordance with the laws of nature. Nabokov reacted against this idea. He clearly felt that it is often the merely lucky, not necessarily the best strategists, who survive.
I thank Jim Crutchfield his many suggestions for ways in which to improve the argument; Kurt Johnson for providing me with a number of helpful lepidoptery facts; and Angus Fletcher for encouraging me to begin investigating this subject. Partial support of this work comes from the Dactyl Foundation for the Arts & Humanities and the Art & Science Laboratory as well as from the Santa Fe Institute, where I was able to formulate this thesis as a visiting researcher January – April 2001.