Hen’s Teeth and Horse’s Toes

by Stephen Jay Gould

  He collected, for example, a large number of geographical races within the species of common wheat, Triticum vulgare. These varied in complex sets of traits, including color of the ears and seeds, form of the ears (bearded or beardless, smooth or hairy), and season of maturation. Vavilov was then surprised and delighted to find virtually the same combinations of characters in varieties of two closely related species, T. compactum and T. spelta.

  He then studied rye (Secale cereale), a species in a genus closely related to wheat but previously regarded as much more limited in its geographical variation. Yet, as Vavilov and his assistants collected rye throughout European and Asiatic Russia, Iran, and Afghanistan, he found not only that its differentiation matched wheat in extent but also that its races displayed the same sets of characters, with the same variations in color, form, and timing of growth.

  The similarities in series of races were so precise and complete between related species that Vavilov felt he could predict the existence of undiscovered varieties within one species after finding their parallel forms in another species. In 1916, for example, he found several varieties of wheat without ligules in Afghanistan (ligules are thin membranes that grow from the base of the leaf blade and surround the stem in many grasses). This discovery suggested that varieties of rye without ligules should also exist, and he grew them from seeds collected in Pamir in 1918. He predicted that durum wheat (Triticum durum), then represented exclusively by spring varieties, should also have winter forms since related species do—and he found them in 1918 in an isolated region of northern Iran.

  Vavilov’s observation would have engendered less controversy had he not interpreted it, indeed overinterpreted it, in a manner uncongenial with strictly Darwinian or Lamarckian views. He might have argued that his series of parallel varieties represented similar adaptations of different genetic systems to common environments that engendered natural selection in the same direction. Such an interpretation would have satisfied Darwinian preferences for random variation, with evolutionary change imposed by natural selection. (It could also have been distorted by Lysenko into a claim that environments directly altered the heredity of plants in favorable ways.)

  But Vavilov proposed a different explanation more in tune with non-Darwinian (though not anti-Darwinian) themes still popular during the 1920s: he claimed that the parallel series of varieties represented identical responses of the same genetic systems, inherited in toto from species to related species. Thus, in evolutionary parlance, his series were “homologous”—hence the name of his law. (Homologies are similarities based on inheritance of the same genes or structures from a common ancestor. Similarities forged within different genetic systems by selective pressures of similar environments are called analogies.)

  Vavilov argued that new species arise by developing genetic differences that preclude interbreeding with related species. But the new species is not genetically distinct from its ancestor in all ways. Most of the ancestor’s genetic system remains intact; only a limited number of genes are altered. The parallel varieties, then, represent a “playing out” of the same genetic capacities inherited as blocks from species to related species.

  Such an interpretation is not anti-Darwinian because it does not deny an important role to natural selection. While each variety may represent a predictable latent capacity, its expression in any climate or geographical region still requires selection to preserve the adaptive variant and to eliminate others. But such an explanation does conflict with the spirit of strict Darwinism because it weakens or compromises the cardinal tenet that selection is the creative force in evolution. Random, or undirected, variation plays a crucial role in the Darwinian system because it establishes the centrality of selection by guaranteeing that evolutionary change cannot be ascribed to variation itself. Variation is only raw material. It arises in all directions or, at least, is not preferentially ordered in adaptive ways. Hence, direction is imposed by natural selection, slowly preserving and accumulating, generation after generation, the variations that render organisms better adapted to local environments.

  But what if variation is not fortuitous and undirected but strongly channeled along certain paths? Then only a limited number of changes are possible, and they record the “internal” constraints of inheritance as much as the action of selection. Selection is not dormant; it still determines which of several possibilities reaches expression in any one climate or area. But if possibilities are strictly limited, and if a species displays all of them among its several varieties, then this range of form cannot be ascribed only to selection acting upon fortuitous variation.

  Moreover, this explanation for new varieties compromises the cardinal principle of creativity for natural selection. The variations are predictable results within their genetic system. Their occurrence is almost foreordained. The role of natural selection is negative. It is an executioner only. It eliminates the variants unfit in any given environment, thus preserving the favored form that had to arise eventually.

  Vavilov interpreted his law of homologous series in this non-Darwinian manner. “Variation,” he wrote, “does not take place in all directions, by chance and without order, but in distinct systems and classes analogous to those of crystallography and chemistry. The same great divisions [of organisms] into orders and classes manifest regularities and repetitions of systems.” He cites the case of “several varieties of vetches so similar to ordinary lentils in the shape, color, and size of their seeds, that they cannot be separated by any sorting machine.” He agrees that the extreme similarity in any one spot is a product of selection—unconscious selection in agricultural sorting machines. But the agent of selection was, literally in this case, only a sieve that preserved one variant among many. The proper variant existed already as a realized product of an inherited set of possibilities.

  The role of natural selection in this case is quite clear. Man unconsciously, year after year, by his sorting machines separated varieties of vetches similar to lentils in size and form of seeds, and ripening simultaneously with lentils. The same varieties certainly existed long before selection itself, and the appearance of their series, irrespective of any selection, was in accordance with the laws of variation.

  Vavilov, overenthused with his own idea, went on. He became intoxicated with the notion that his law might represent a principle of ordering that would render biology as exact and experimental as the “hard” sciences of physics and chemistry. Perhaps genetic systems are composed of “elements.” Perhaps the geographical varieties of species are predictable “compounds” that arise inevitably from the union of these elements in specified mixtures. If so, the ranges of biological form within a species might be laid out in a table of possibilities similar to the periodic table of chemical elements. Evolution might be deduced from genetic structure itself; environment can only act to preserve inherent possibilities.

  He advocated an explicit “analogy with chemistry” in the concluding section of his 1922 paper and wrote: “New forms have to fill vacancies in a system.” He experimented with a style of notation that expressed varieties of a species as a chemical formula and advocated “the analogy of homological series of plants and animals with systems and classes of crystallography with definite chemical structures.” One zealous supporter commented that “biology has found its Mendeleev.”

  Vavilov moderated his views during the 1920s and early 1930s. He learned that some of the parallel varieties between species are not based upon the same genes after all, but represent the similar action of selection upon different sources of variation. In these cases, the varieties are analogous, not homologous, and the Darwinian explanation must be preferred. He wrote in 1937:

  We underestimated the variability of the genes themselves…. At the time we thought that the genes possessed by close species were identical; now we know that this is far from the case, that even very closely related species which have externally similar traits are characterized by many different genes. By concentrating our
attention on the variability itself, we gave insufficient attention to the role of selection.

  Still, Vavilov continued to champion the importance and validity of his law, and he continued to advocate the analogy with chemistry in only slightly weakened form.

  Unfortunately, in the deepest sense, Vavilov had left himself open to Lysenko’s polemical attack. The law of homologous series provided Lysenko with important ammunition, and Vavilov’s overextended chemical analogy deepened his troubles. Lysenko caricatured Vavilov’s law in 1936 by presenting ridiculous examples involving species too distantly related to present parallel series in Vavilov’s system: “In nature we find apple trees with round fruit, hence there must or can be trees with round pears, cherries, grapes, etc.”

  Lysenko’s ideological attack was more vicious. He made two major charges involving both parts of that catchword for official Soviet philosophy—dialectical materialism. Vavilov’s law, he claimed, was undialectical because it located the source of organic change within the genetic systems of organisms themselves and not in the interaction (or dialectic) between organism and environment. Secondly, Lysenko charged that the law of homologous series was “idealist” rather than materialist because it viewed the evolutionary history of a species as prefigured in the unrealized (and therefore nonmaterial) capacity of an inherited genetic system.

  Evolution, Lysenko charged, is almost an illusion in Vavilov’s scheme. It represents a mere playing out of inherited potentials, not the development of anything new. It expresses the bourgeois penchant for stability by depicting apparent change as a superficial expression of underlying constancy. According to Vavilov’s law, Lysenko charged,

  New forms result not from the development of old forms, but from a reshuffling, a recombination of already existing hereditary corpuscles…. All the existing species existed in the past, only in less diverse forms; but every form was richer in potentialities, in its collection of genes.

  Madness often displays a perverse but cogent reason in its own terms; and we must admit that Lysenko did identify and exploit the true weaknesses in Vavilov’s argument. Vavilov did underplay the creative role of environment, and his chemical analogy did betray a belief in prefigured potentiality as the source of later, and in some sense illusory, change. But Lysenko, who was also both a charlatan and a cruel polemicist, was equally undialectical (despite his protestations to the contrary) in viewing plants as putty before a molding environment.

  Vavilov died in the name of a phony Lamarckism. He became a legitimate martyr in the West, but his ideas did not flourish as a result. The law of homologous series, the organizing theme of his evolutionary work, was ignored in the name of an overly strict Darwinism. Vavilov’s law did not directly contradict Darwinian principles, but its emphasis on constraints of inheritance and channeled variation fit poorly with the favored Darwinian theme of random variation and the guidance of evolutionary change by natural selection. It was therefore neglected and relegated to the shelf of antiquated theories that had implicated variation itself as a directing force in evolution. I have consulted all the founding documents of the “modern synthesis,” the movement that established our present version of Darwinism between the late 1930s and the 1950s. Only two mention Vavilov’s law of homologous series, each in less than one paragraph.

  Yet I feel that in his imperfect way Vavilov had glimpsed something important. In more modern terms, new species do not inherit an adult form from their ancestors. They receive a complex genetic system and a set of developmental pathways for translating genetic products through embryology and later growth into adult organisms. These pathways do constrain the expression of genetic variation; they do channel it along certain lines. Natural selection may choose any spot along the line, but it may not be able to move a species off the line—for selection can only act upon the variation presented to it. In this sense, constraints of variation may direct the paths of evolutionary change as much as selection acting in its Darwinian role as a creative force.

  I have found Vavilov’s views very helpful in reorienting my own thinking in directions I regard as more fruitful than my previous unquestioned conviction that selection manufactures almost every evolutionary change. In studying the relationship of brain size to body size, biologists find that brains increase only one-fifth to two-fifths as fast as bodies in comparisons of closely related mammals differing only (or primarily) in body size—adults within a single species, breeds of domestic dogs, chimpanzees versus gorillas, for example. For ninety years, the large literature has centered on speculations about the adaptive reasons for this relationship, based upon the (usually unstated) assumption that it must arise as the direct product of natural selection.

  But my colleague Russell Lande recently called my attention to several experiments on mice selected over several generations for larger body size alone. As these mice increased in size across generations, their brains enlarged at the characteristic rate—a bit more than one-fifth as fast as body size. Since we know that these experiments included no selection upon brain size, the one-fifth rate must be a side product of selected increase in body size alone. Since the one-fifth to two-fifths rate appears again and again in diverse lineages of mammals, and since it may record a nonadaptive response of brains to selection for larger bodies within mammalian developmental systems, the parallel sets of races and species arrayed along the one-fifth to two-fifths slope in carnivores, rodents, ungulates, and primates are non-Darwinian homologous series in Vavilov’s sense.

  In personal research on the West Indian land snail Cerion, my colleague David Woodruff and I find the same two morphologies again and again in all the northern islands of the Bahamas. Ribby, white or solid-colored, thick, and roughly rectangular shells inhabit rocky coasts at the edges of banks where islands drop abruptly into deep seas. Smooth, mottled, thinner, and barrel-shaped shells inhabit calmer and lower coasts at the interior edges of banks, where islands cede to miles of shallow water. The easiest, and usual, conclusion would view ribby shells on all islands as closely related and smooth shells as members of another coherent group. But we believe that the complex set of characters forming the ribby and smooth morphologies arise independently, again and again. On the islands of Little Bahama Bank, both ribby and smooth animals share a distinctive genital anatomy. On the islands of Great Bahama Bank, both ribby and smooth develop an equally distinctive, but different, kind of penis. The ecology of rocky versus calm coasts may select for ribby and smooth morphologies as adaptations, but the coordinated appearance of the half dozen distinctive traits of each morphology may represent a channeling of available variation to produce homologous series (ribby and smooth varieties) in different lineages (defined by genital anatomy).

  A complete theory of evolution must acknowledge a balance between “external” forces of environment imposing selection for local adaptation and “internal” forces representing constraints of inheritance and development. Vavilov placed too much emphasis on internal constraints and downgraded the power of selection. But Western Darwinians have erred equally in practically ignoring (while acknowledging in theory) the limits placed upon selection by structure and development—what Vavilov and the older biologists would have called “laws of form.” We need, in short, a real dialectic between the external and internal factors of evolution.

  Vavilov’s personal tragedy cannot be undone. But he has been rehabilitated in Russia, where the All-Union Society of Geneticists and Selectionists now bears his name. We who view him as a martyr and champion his name while ignoring his ideas would do well to reconsider the older non-Darwinian tradition that he represented. Combined with our legitimate conviction about the power of selection, the principle of homologous series (and other “laws of form”) might foster an evolutionary theory truly synthetic in its integration of development and organic form with a body of principles now dominated by ecology and the effects of selection upon single genes and traits.

  3 | Adaptation and Development

  11 | Hy
ena Myths and Realities

  I FREELY ADMIT that the spotted, or laughing, hyena is not the loveliest animal to behold. Still, it scarcely deserved the poor reputation imposed upon it by our illustrious forebears. Three myths about hyenas helped to inspire the loathing commentary of ancient texts.

  Hyenas, first of all, were regarded as scavengers and consumers of carrion. In his Natural History, Pliny the Elder (A.D. 23–79) spoke of them as the only animals that dig up graves in search of corpses (ab uno animali sepulchra erui inquisitione corporum). Conrad Gesner, the great sixteenth-century cataloger of natural history, reported that they gorge themselves so gluttonously after finding a corpse that their bellies swell to become taut as a drum. They then seek a narrow place between two trees or stones, force themselves through it, and extrude the remains of their meal simultaneously at both ends.

  Hans Kruuk, who spent years studying spotted hyenas on their home turf (the plains of East Africa), has labored to dispel these ancient myths (see his book The Spotted Hyena, University of Chicago Press, 1972). He reports that hyenas will scavenge when they get the opportunity. (Almost all carnivores, including the noble lion, will happily feast upon the dead product of another animal’s labor.) But spotted hyenas live in hunting clans of up to eighty animals. Each clan controls a territory and kills most of its own food—mainly zebra and wildebeest—in communal, nocturnal pursuit.