Nabokov and the Species Concept
When »Carl von Linné devised his system of nature, nobody thought of evolution. He was a pious man and believed he was merely uncovering the divine plan, neatly arranging into a system what a supreme artificer had wisely and orderly created. If it presented itself as a chaotic jumble of forms, that was only due to man's unfortunate lack of understanding. All species, he and his coevals believed, had been created within a week and had stayed the same ever since. Species were something fixed, immutable, never-changing. They were all of the same age, and new species did not arise. There were no blurred borders between them. Each was the embodiment of a different divine idea, so to speak. Behind every given specimen there lurked the 'type' which was actually Infinite Wisdom's plan for the species it belonged to. True, there were variations within many of them. They did not disturb early naturalists too much who dismissed them as alterations of the type that had been brought about by some action of the environment. This is the origin of what was later called the 'typological' species concept, founded on religion. As the lepidopterist, Charles »Oberthür, put it at the end of the nineteenth century, taxonomy is "the knowledge of the classification according to which the Supreme Intelligence has grouped and ordered the species."
Only a century after Linné it turned out that the differing degrees of morphological similarity his system was based on were due to the fact that organisms were in fact related, not only morphologically, but phylogenetically. They were similar because they were related to each other and descended from one another. The manifold forms of nature are the result of that slow and steady branching process known as evolution. Without knowing it, the system of nature Linné and the naturalists in his wake had been constructing was a phylogenetic one.
For systematics, this brought a change. Overall similarity was not enough to place any two species in the same genus, or any two genera in the same family. Similarities, in fact, may be quite misleading. Ideally, for taxonomy to reflect phylogeny, all members of each unit of classification (called a taxon) should be descended from one ancestor. The word is 'monophyletic.' Every taxon should be monophyletic. Every species of a genus should be descended from one early ancestor, as every family of genera should be descended from only one still earlier form. The problem is that it is not easy to determine whether any one group is monophyletic. If different features are examined (say, wing venation, the shape of the antennae or the knobs on the caterpillar), conflicting hypothetical lineages may ensue.
When what is visible to the naked eye did not seem conclusive enough any more, entomology turned to anatomical dissection under the microscope. As a result it became evident that many species which had been placed in the same genus because of their over-all similarity could not be descended from the same ancestral form. In these cases it was time for a nomenclatorial change. The whole genus and the related genera had to be reexamined and reevaluated. Sometimes with a lot of reshuffling, the whole bunch then was rearranged. There has been a bewildering amount of renaming in lepidoptery.
Still, the typological concept of the species held on far into the twentieth century. In his younger years, Nabokov had witnessed it crumbling. The conventional "philatelistic" approach to lepidoptery had been toppled. Early in the century, Nabokov wrote in his memoir, there was a change "which coincided with my ardent adolescent interest in butterflies and moths. The Victorian and Staudingerian kind of species, hermetic and homogeneous, with sundry (alpine, polar, insular, etc.) 'varieties' affixed to it from the outside, as it were, like incidental appendages, was replaced by a new, multiform and fluid kind of species, organically consisting of geographical races or subspecies. The evolutional aspects of the case were thus brought out more clearly, by means of more flexible methods of classification, and further links between butterflies and the central problems of nature were provided by biological investigations."
So Nabokov had left the rigid typological species concept behind, exchanging it for one that recognized the unstable, fluid nature of the species and paved the way to evolutionary considerations. As Nabokov used the word himself, let us call this species concept 'organic.'
Shortly before Nabokov came to the United States, he made Fyodor's father (in "Father's Butterflies") disparage genitalic dissection: "… he subtly berated those whom he called 'genitalists': it was just the time when it became fashionable to accept as an unerring and adequate sign of species differentiation distinctions in the chitinoid structure of the male organ, which represented, as it were, the 'skeleton' of a species, a kind of 'vertebra.'" There is no knowing whether Nabokov was voicing his own opinions of the time or only those Godunov Senior might very well have held around 1917. Anyway, as soon as he convinced himself how successfully American lepidopterists like William P. Comstock at the American Museum of Natural History were practicing genitalic dissection, he learned the methods and became an ardent partisan and an expert in this kind of structural analysis himself. Nabokov emphatically sided with the innovators.
It so happened that in the forties another paradigm shift was in the works, one that was to revolutionize all of biology. It was the Neo-Darwinian turn merging Mendelian genetics and the Darwinian explanation of evolution through mutation and natural selection. This time Nabokov did not side with the new. Just before he came to America, he had tentatively elaborated his anti-Darwinian feelings in the essay "Father's Butterflies" written to be appended to his novel The Gift. He never finished let alone published it, but the feeling seems to have remained with him.
The biological species concept that evolved along with the Neo-Darwinian synthesis was not a necessary corollary but emerged from the same quarters. The common-sense view that had prevailed through the centuries was that a species simply is any group of organisms that look very much alike. One species was told from the other solely on morphological grounds. Obvious morphological characteristics in Lepidoptera are overall pattern, coloring and size. Nabokov certainly was not one of these naïve typologists when he came to United States. He did not consider superficial morphological characteristics very helpful in making finer distinctions and soon came to emphasize the importance of microscopic features, especially the genitalic structures. Yet he seems to have been worried that too much biology would dethrone morphology.
In his paper on the morphology of the genus Lycaeides (Lep8, 1944), Nabokov also introduced the scale line count. Like the tiles on a roof, the scales are placed in lines. These lines, Nabokov found, form "concentric rings or ripples…, radiating from a center more or less coincident with the base of the wing" (Lep14 483, 1949). He used the number of lines to describe the exact position of various wing markings and macules, probably hoping to thus gain another set of facts which would help him in the taxonomic delineation of the species and subspecies of the genus. However, in his definite paper on the North American members of Lycaeides (Lep14, 1949), while citing scale line counts, he did not use them for taxonomic purposes but stuck to genitalic structures. There was just too much overlap in the variability of wing characters between the two species concerned, idas and melissa. To use the term coined by Nabokov (Lep8 137, 1944), there was just two much 'homopsis,' too much striking similarity of the wing characters. "… certain specific tendencies can be distinguished, despite the fact that both species go through much the same racial aspects (from pale, poorly marked, to dark, strongly marked forms)" (Lep14 541, 1949).
As Ernst »Mayr put it, one of the eminent evolutionists who pioneered the new biological view of the species, "A concept based on the degree of morphological distinction is the typological species concept of the old systematics and is even today the only practical one in all those systematic groups which are still in the descriptive or cataloguing stage. It is the species concept with which Linnaeus started the science of systematics in his Systema naturae … The characters which the taxonomist enumerates in his species diagnosis, such as structure, proportions, color patterns, and the like, are the conventional criteria used in the old systematics to define a species. 'A species is a group of individuals or populations with the same or similar morphological characters.'"
In contrast, this is how Mayr himself defined the new biological species concept: "Species are groups of actually or potentially interbreeding natural populations, which are reproductively isolated from other such groups."
The old typological species concept did not suffice for several reasons. Species are not fixed and rigid; they consist of peer variant forms none of which can be said to be the ideal type. They may have blurry borders, one shading into the other so it is impossible to decide where one ends and the next begins. They are not stable over time – if you watch close enough, you may even see the gene pool of a population change from generation to generation. Often there is more variation within a species than between two species. The blue butterflies that comprise the idas group within the genus Plebejus are a case in point (idas, argyrognomon, melissa, subsolanus, etc.). Very superficially, they all are very much alike. If you saw them fluttering about a puddle, you would be hard pressed to say exactly what species they are. Yet there would be differences, only that they may not run along species borders. How are taxonomists to know which is which? They have to resort to a microscopic examination of the sclerotized genitalia to tell the difference, and they will be glad to have found a characteristic that is both telling and invariant enough. Yet even the microscopic examination may not settle the question, and morphology will have exhausted its resources and be at an end. Could the question then never be decided, not even in theory?
That is why abandoning the typological concept based on morphology alone promised several advantages. According to Mayr,
(1) the biological species concept allows for a non-arbitrary, objective definition of the species;
(2) it will ultimately allow to define a species not only on account of its visible features but on account of its genetic relationship to other species;
(3) most importantly, it does demonstrate not only that there are similarities and dissimilarities, but "why there should be any."
That is, the biological species concept links taxonomy to all of evolutionary biology, by considering the species as a genetic unit (a certain gene pool sustained by a flow of genes between the individuals) acted upon by mutation and selection, the two moving forces of evolution. It provides an explanation of why taxonomy reflects phylogeny. After centuries of descriptive inventorying, this indeed is the story biology has begun to unfold.
'Population' was a concept Nabokov was wary of. He thought it "a dreadfully misused term – and a hideous word, anyway" (»Lep8, 1944). Still it is here to stay. The biological view defines a population as a group of organisms capable of interbreeding and coexisting in the same place and at the same time. As long as the members of a population keep interbreeding, they swap their genes, and a gene-flow is maintained. If there is an advantageous mutation, it may spread to other members of the population; a deleterious mutation will not spread but delete its bearer. When a population is split and separated, the gene-flow ceases. For a time the two populations could still interbreed if they were joined again. Sooner or later so many mutations will accumulate in either that they could no longer interbreed successfully. At this point the link snaps, and they become distinct species. Members of different species cannot interbreed, no matter how much they may resemble one another.
In reality, of course, things are seldom that clear-cut. Nature is fuzzy. There is a border zone between closely related species where it may be impossible to tell whether they can interbreed or not. Hybridization seems to be a widespread phenomenon among closely related species. Even if two sympatric species normally don't interbreed, there may be some individuals that do. Again normally, the hybrids will be at a disadvantage. Often they will be infertile, like mules. Others may not be able to exploit the niches where their parent species thrived with the same success and consequently will be doomed to disappear. But if conditions change, the old niches shrinking and new ones opening, hybrids may suddenly see themselves at an advantage. They may be better able to cope with the new conditions. In critical situations it pays if there is a good deal of variability in a population. It increases the chance that at least some of its members will have the right variant of body or behavior to come to terms with the hardship. So in microevolution, there are forces that tend to keep related species separate and others that tend to fuse them, and they may oscillate over a long span of time.
This does not invalidate the biological species concept. It is just a footnote to it, saying that new species don't erupt and that speciation may be preceded by long phases of experimentation in nature. While they last, the reproductive barrier will not be an iron curtain. There will be little successful interbreeding between quasi-separate populations, yet ties will not be definitely severed. That is why the biological species concept is not of much help to the immediate concerns of practical-minded taxonomists. They want to know whether two populations are the same species or not, or if they are different subspecies of one species. If they have problems determining their exact relationship, it may be because nature itself has not decided yet.
Nabokov seems to have given the biological species concept much thought during his years at the Harvard Museum of Comparative Zoology but was reluctant to accept it, as is evident from his notes on the subject published in Nabokov's Butterflies. He especially resented Mayr's notion of "potential interbreeding" which he seems to have taken not so much as a theoretical concept but as a practical guideline, absurdly suggesting entomologists should stop looking at the morphology of insects and begin probing whether they can produce interbreeding in the laboratory.
So what Nabokov particularly objected to was the way the new biological view seemed to de-emphasize morphology: "The strictly biological meaning forcibly attached by some modern zoologists to the specific concept has crippled the latter by removing the morphological moment to a secondary or still more negligible position, while employing terms, e.g. 'potential interbreeding,' that might make sense only if an initial morphological approach were presupposed. What I term species, in my department, can be defined as a phase of evolutional structure, male and female, traversed more or less simultaneously by a number of, consequently, more or less similar organisms shading into each other in various individual or racial ways, interbreeding in a given area and separated there from sympatric representatives [that is those sharing the same area] of any other such phase by a structural hiatus with absence of interbreeding between the two sets" (Lep9 3, 1945). Put more briefly, "A species is a relative category, at its tangible best represented by a number of interbreeding organisms which constantly differ in structure from and do not interbreed with any other organism inhabiting the same area." In 1963, he complained, "The enthusiastic reliance on biological data with meagre morphological characters has led to the erection of many very doubtful species."
In short, Nabokov always stressed the supreme importance of morphology but by 1944 had unequivocally accepted the biological view that it is the capability of interbreeding that delimits a species, at least for sympatric populations. As for allopatric populations (that is, those inhabiting different areas), Nabokov went on to believe that the concept of 'potential interbreeding' made no sense. In this, however, he was not far apart from Mayr who wrote, "When some of the more distant populations are geographically isolated from all other populations within a species, the question arises: Are these isolated populations still members of the parental species? What criteria can one use to decide which of these populations to recognize as full species and which others to combine into a polytypic species? The species status of geographically isolated populations can be determined only by inference, particularly by degree of morphological difference." That is, for allopatric populations both Mayr and Nabokov thought morphology offered the only approach possible. As far as the species concept was concerned, there was practically no disagreement.
What Nabokov did not accept, however, was the Neo-Darwinian explanation of how speciation and hence evolution come about. Of course, Nabokov was fully aware that morphological variation was caused by evolution and that taxonomy was spelling out the "phylogenetic sense" of it all. He believed taxonomy should reflect phylogeny and inferred phylogenetic lineages from morphological analysis. However, "while accepting evolution as a modal formula, I am not satisfied with any of the hypotheses advanced in regard to the way it works; on the other hand, I am quite certain that repetitions of structure … cannot be treated as a result of haphazard 'convergence' since the number of coincident characters in one element, let alone the coincidence of that coincident number with a set of characteristics in another element, exceeds anything that might be produced by 'chance'" (»Lep9, 1945).
In other words, he doubted the Darwinian theory that evolution was based on the two-step process of mutation (random alteration at the genetic level) and selection (preservation of mutations that offer reproductive advantages to their bearers). He seems to have judged the theory too utilitarian, leaving no room for the development of any organismic character that was not of some use to its bearer. He did not put forward a methodical argument against Darwinism, but here and there he voiced his doubts, for instance in Speak, Memory: "'Natural selection,' in the Darwinian sense, could not explain the miraculous coincidence of imitative aspect and imitative behavior… I discovered in nature the nonutilitarian delights that I sought in art."
In his technical papers he refrained from affirming and explaining his anti-Darwinian stance. It would have put him in an increasingly awkward position within the scientific community that was almost universally embracing the theory of natural selection. In Arnold Mallis' American Entomologists (1971), there is a chapter on Nathan »Banks, the Head Curator of Insects at the Harvard Museum of Comparative Zoology when Nabokov took up his work there. Quoting Carpenter & Darlington, Mallis says, "It is a misfortune that [Banks'] knowledge of the structural diversity and adaptations of insects was not more available to biologists. This was partly his fault. He had not learned the modern vocabularies of genetics and evolution … He did not join the smoking and conversation group on the Museum steps …" – presumably discussing genetics and evolution. We do not know if and how much Nabokov sympathized with Banks' aloofness. Certainly he did not join the buzz either.
It should be emphasized, however, that Nabokov's rejection of Darwinism in no way impeded or tainted his scientific work as a taxonomist. Taxonomy did and for the most part still does have to rely on morphology. Nabokov explicitly wanted classification to reflect "the structural relationship" and the "phylogenetic circumstances." For all practical purposes, this was sufficient to ensure that his taxonomic work did not lag behind the times.
Knowing how much he disagreed with Darwinism, one is struck by an otherwise innocuous sentence in his last scientific statement (Lep22): "For better or worse our present notion of species in Lepidoptera is based solely on the checkable structures of dead specimens, and if Forster's Furry cannot be distinguished from the Furry Blue except by its chromosome number, Forster's Furry must be scrapped". It is a statement that merits a closer look.
Forster's Furry was considered a subspecies of the Furry Blue and hence belongs to the group of butterflies Nabokov studied most closely. Its scientific name is Polyommatus (formerly Agrodiaetus) ainsae Forster, 1961, the name deriving from the Spanish town of Ainsa at the southern foot of the Pyrenees where the little blue was first found. It was established as a subspecies of Polyommatus dolus by the Munich entomologist Walter Forster in his revision of the genus Agrodiaetus Scudder. (Later, in 1997, the genus Agrodiaetus was swallowed up by the larger genus Polyommatus.) Nabokov found it listed in Higgins' and Riley's Field Guide to the Butterflies in Great Britain and Europe which he reviewed right after its publication in 1970. The entry carried the note, "A. ainsae is distinguished specifically from A. dolus by its lower chromosome number."
A look at »Forster's Original Description shows, however, that the taxon ainsae had been established on purely morphological grounds, and on the most hackneyed and external "Staudingerian" ones at that. Forster had promoted what had formerly been considered a mere aberration of Polyommatus dolus vittata Oberthür, 1892 to the rank of a subspecies because it was smaller than dolus, because its blue was lighter, because there was a white blot on the underside of its hindwings which dolus lacked and because it seemed to be occurring regularly in the region of Ainsa and Jaca. There was no mention of the chromosome count at all. In the introduction to one of Forster's previous papers there was just a reference to "the highly interesting cytological studies of H. de Lesse in Paris the results of which will soon be published."
What would a different chromosome count mean? Butterflies are unique in that they may sometimes interbreed successfully even with slightly differing numbers of chromosomes. It has been shown that there are species where the exact number of chromosomes does not seem to matter. This is because in these species the parents' chromosomes do not have to be matched one by one like they have to elsewhere in nature, with havoc occurring when some are left over. Instead, the superfluous ones are simply ignored and discarded. Still, in general it is true also for butterflies and moths that if two individuals have a different karyotype (that is, if their chromosomes differ in number and form), this is as strong an indication as any that they will not be able to mate successfully. Hence they would belong to different species, no matter how much alike they look. For this reason the karyotype has been used as additional evidence when distinguishing different taxa.
The search for ainsae's chromosome count turns out to be a paper chase. There is no hint in the Higgins & Riley Field Guide as to where the authors found the information. Five years later, one of them published a strictly scientific book on European butterflies, and there he says about Agrodiaetus, "The numerous species, which are distributed throughout the Mediterranean subregion, include forms almost or quite indistinguishable by external characters or genitalia, but with differences in chromosomal numbers (karyotype) in geographically localised races, which are accepted here with specific rank." As the only source he cites the work of the French entomologist to whose forthcoming publication Forster had pointed fifteen years before. This, however, was not Hubert de Lesse's first and fundamental paper on the chromosomes of butterflies which had appeared in 1960 but a later one, and it had nothing on ainsae. The information concerning ainsae is in one of the supplementary later papers by de Lesse. While dolus dolus north of the Pyrenees had a rather invariant chromosome count of 124, dolus ainsae had an equally invariant one of 108. So de Lesse's cytological work had supported Forster's conclusion from morphology that dolus dolus and dolus ainsae are different taxa, without establishing whether ainsae was a subspecies of dolus or a different species. Today it is considered a species.
Curiously, even a taxonomist insisting on the supreme importance of morphology would not have to dismiss the chromosome count at all. The karyotype in a way is not less of a structural characteristic than the uncus of a male butterfly. Differences in the form and number of the chromosomes can be seen under a microscope, just like the differences in genitalia. All one needs is a stronger microscope. Limiting the examination of structures to the resolution of the optical microscope would be raising a technical obstacle to the status of an ontological criterion.
So why should Nabokov have wanted to scrap ainsae at all? Perhaps he wouldn't have had he still been in his lab and actually looked at the chromosomes through one of the more powerful microscopes. One can only speculate that it was because the chromosome count is a very gross intimation of a creature's genetic make-up and because biology, in the framework of the theory of natural selection, had become very much interested in that. The relatedness of living beings is coded in their genetic make-up. From the chromosome count biology proceeded to much subtler methods of determining the genetic make-up, and this indeed may in some respects change the ways of taxonomy. To such tools as tweezers and microscopes zoologists have added some of the tools of molecular biology used for DNA mapping. As DNA deteriorates fast, dead museum specimens ("the checkable structures of dead specimens") will not do. The researcher needs some live DNA from the insects to be tested, so he may not chloroform or pinch them to death in the field but has to bring them to his lab alive.
With proteins called restriction enzymes, the DNA strand is cut at specific sequences called recognition sites. The ensuing fragments are then separated according to length by electrophoresis. Placed in an agarose gel to which an electric field is applied, the negatively charged fragments will migrate the faster down towards the positive end the smaller and lighter they are. If the process is stopped at the right moment, the fragments of different length will be neatly deployed over the whole gel. Now if there have been mutations, it is likely that some of the recognition sites will be involved and that consequently the same restriction enzyme will chop the mutated DNA strand into different fragments. To determine the genetic distance between two taxa, the scientist selects a number of telling DNA fragments that are more or less stable across the first species. Their being stable means that there are no or few mutations in these sequences. Then he tests how these fragments compare to the ones the same restriction enzyme produces in the other species. The fewer of the stable fragments are repeated, the more mutations must have accrued to separate the second species from the first. The method thus detects genetic polymorphisms, and their number can be used to estimate the genetic distance between the two taxa. If ainsae were genetically undistinguishable from dolus, it would have to be "scrapped" on biomolecular grounds. If it could be clearly distinguished, DNA mapping would have provided good further reason to consider it a species of its own.
The perennial question of subspecies versus species and other urgent questions of taxonomy will not be resolved by DNA mapping, though. There simply are no generally agreed on thresholds beyond which genetic dissimilarity adds up to the formation of either a new species or a subspecies (like "a .25 per cent difference makes a subspecies and a .5 per cent one a species"), and there never will be. If mutations hit the right sequences, very few of them may be enough to prevent two populations from mating successfully. In a paper analyzing 103 characters in 59 exemplar species for the purpose of higher (that is, supergeneric) classification, the authors on the one hand concluded that adding more morphological data would hardly give better results and that molecular studies are likely to yield more useful information. On the other hand, "initial hopes that DNA analysis would provide a clear and unequivocal solution to all systematic problems can now be seen as overoptimistic. … Molecular data are not intrinsically better; we should not rely on biochemistry alone, but look for consensus amongst results obtained from molecular and morphological data sets." It seems that the rivalry between morphological and molecular methods is giving way to a symbiosis.
For even if molecular biology will not settle all the problems of taxonomy, it may be helpful. For instance, a DNA mapping study in 1999 confirmed Nabokov's conclusion (»Lep14, 1949) that Plebejus idas and Plebejus melissa are indeed different species. Molecular genetics turned out to be in agreement with morphology. Kurt Johnson tells that the copy of the paper one of the authors sent him has the comment, "Do you think Vladimir's Shade is pleased? I do." In 2011, a further DNA study finally established that the »Karner Blue is a species in its own right. To Nabokov, that would have been wonderful news.
I like to imagine that Nabokov would finally have made his peace with the theory of natural selection and with the methods of investigating the genetic make-up. It may be noted that when in 1968 his one-time adversary Ernst Mayr wrote to him in Montreux, he got a particularly warm reply. And in one of his last interviews, Nabokov cited Darwin as the epitome of the "true scientist", a man of "acute imagination".
 Speak, Memory, p. 124
 Nabokov's Butterflies, p. 208
 Systematics and the Origin of Species, 1942
 Nabokov's Butterflies, p. 340
 Nabokov's Butterflies, p. 578
 Ernst Mayr, This is Biology, Cambridge, Massachusetts (Harvard UP) 1997, p.130
 Speak, Memory, p. 125
 Strong Opinions, p. 320
 "Bausteine zur Kenntnis der Gattung Agrodiaetus Scudd. (Lep. Lycaen.) II," Zeitschrift der Wiener Entomologischen Gesellschaft, 46 (5), 1961, p. 74–9
 Walter Forster, "Einige neue Formen der Gattung Agrodiaetus Scudd. (Lep. Lycaen.)," Entomologische Zeitschrift (Stuttgart), 70 (3), 1. Februar 1960, p. 17–22
 Lionel G. Higgins, The Classification of European Butterflies, London (Collins) 1975, p. 155
 H. de Lesse, "Variation chromosomique chez Agrodiaetus dolus Hübner," Annales de la Société entomologique de France (Paris), 2 (1), 1966, p. 209–14
 "Variation chromosomique chez Agrodiaetus dolus Hb.," Alexanor (Paris), 2, 1962, p. 283–86
 Riek de Jong / Richard I. Vane-Wright / Phillip R. Ackery, "The higher classification of butterflies (Lepidoptera): problems and prospects," Entomologica scandinavica (Copenhagen), 27 (1), 1996, p. 65–101
 Chris C. Nice / Arthur M. Shapiro, "Molecular and morphological divergence in the butterfly genus Lycaeides [Lepidoptera: Lycaenidae] in North America: evidence of recent speciation," Journal of Evolutionary Biology (Oxford), 12, 1999, p. 936–50
 In the listserve Nabokv-L, November 1999
 Nabokov's Butterflies, p. 651
 Mati Laansoo, "An interview with Vladimir Nabokov for the CBC," The VN Research Newsletter (Lawrence, Kansas), 10, Spring 1983, p. 44