Hen’s Teeth and Horse’s Toes

by Stephen Jay Gould

  Second, in forming and churning the soil, they maintain a steady state amidst constant change. As the primary theme of his book (and the source of its title), Darwin set out to prove that worms form the soil’s upper layer, the so-called vegetable mold. He describes it in the opening paragraph:

  The share which worms have taken in the formation of the layer of vegetable mould, which covers the whole surface of the land in every moderately humid country, is the subject of the present volume. This mould is generally of a blackish color and a few inches in thickness. In different districts it differs but little in appearance, although it may rest on various subsoils. The uniform fineness of the particles of which it is composed is one of its chief characteristic features.

  Darwin argues that earthworms form vegetable mold by bringing “a large quantity of fine earth” to the surface and depositing it there in the form of castings. (Worms continually pass soil through their intestinal canals, extract anything they can use for food, and “cast” the rest; the rejected material is not feces but primarily soil particles, reduced in average size by trituration and with some organic matter removed.) The castings, originally spiral in form and composed of fine particles, are then disaggregated by wind and water, and spread out to form vegetable mold. “I was thus led to conclude,” Darwin writes, “that all the vegetable mould over the whole country has passed many times through, and will again pass many times through, the intestinal canals of worms.”

  The mold doesn’t continually thicken after its formation, for it is compacted by pressure into more solid layers a few inches below the surface. Darwin’s theme here is not directional alteration, but continuous change within apparent constancy. Vegetable mold is always the same, yet always changing. Each particle cycles through the system, beginning at the surface in a casting, spreading out, and then working its way down as worms deposit new castings above; but the mold itself is not altered. It may retain the same thickness and character while all its particles cycle. Thus, a system that seems to us stable, perhaps even immutable, is maintained by constant turmoil. We who lack an appreciation of history and have so little feel for the aggregated importance of small but continuous change scarcely realize that the very ground is being swept from beneath our feet; it is alive and constantly churning.

  Darwin uses two major types of arguments to convince us that worms form the vegetable mold. He first proves that worms are sufficiently numerous and widely spread in space and depth to do the job. He demonstrates “what a vast number of worms live unseen by us beneath our feet”—some 53, 767 per acre (or 356 pounds of worms) in good British soil. He then gathers evidence from informants throughout the world to argue that worms are far more widely distributed, and in a greater range of apparently unfavorable environments, than we usually imagine. He digs to see how deeply they extend into the soil, and cuts one in two at fifty-five inches, although others report worms at eight feet down or more.

  With plausibility established, he now seeks direct evidence for constant cycling of vegetable mold at the earth’s surface. Considering both sides of the issue, he studies the foundering of objects into the soil as new castings pile up above them, and he collects and weighs the castings themselves to determine the rate of cycling.

  Darwin was particularly impressed by the evenness and uniformity of foundering for objects that had once lain together at the surface. He sought fields that, twenty years or more before, had been strewn with objects of substantial size—burned coals, rubble from the demolition of a building, rocks collected from the plowing of a neighboring field. He trenched these fields and found, to his delight, that the objects still formed a clear layer, parallel to the surface but now several inches below it and covered with vegetable mold made entirely of fine particles. “The straightness and regularity of the lines formed by the embedded objects, and their parallelism with the surface of the land, are the most striking features of the case,” he wrote. Nothing could beat worms for a slow and meticulous uniformity of action.

  Darwin studied the sinking of “Druidical stones” at Stonehenge and the foundering of Roman bathhouses, but he found his most persuasive example at home, in his own field, last plowed in 1841:

  For several years it was clothed with an extremely scant vegetation, and was so thickly covered with small and large flints (some of them half as large as a child’s head) that the field was always called by my sons “the stony field.” When they ran down the slope the stones clattered together. I remember doubting whether I should live to see these larger flints covered with vegetable mould and turf. But the smaller stones disappeared before many years had elapsed, as did every one of the larger ones after a time; so that after thirty years (1871) a horse could gallop over the compact turf from one end of the field to the other, and not strike a single stone with his shoes. To anyone who remembered the appearance of the field in 1842, the transformation was wonderful. This was certainly the work of the worms.

  Section through one of the fallen Druidical stones at Stonehenge, showing how much it had sunk into the ground. Scale ½ inch to 1 foot.

  An original illustration from Darwin’s worm book showing the foundering of large stones by the action of worms.

  In 1871, he cut a trench in his field and found 2.5 inches of vegetable mold, entirely free from flints: “Beneath this lay coarse clayey earth full of flints, like that in any of the neighboring ploughed fields…. The average rate of accumulation of the mould during the whole thirty years was only .083 inch per year (i.e., nearly one inch in twelve years).”

  In various attempts to collect and weigh castings directly, Darwin estimated from 7.6 to 18.1 tons per acre per year. Spread out evenly upon the surface, he calculated that from 0.8 to 2.2 inches of mold would form anew every ten years. In gathering these figures, Darwin relied upon that great, unsung, and so characteristically British institution—the corps of zealous amateurs in natural history, ready to endure any privation for a precious fact. I was particularly impressed by one anonymous contributor: “A lady,” Darwin tells us, “on whose accuracy I can implicitly rely, offered to collect during a year all the castings thrown up on two separate square yards, near Leith Hill Place, in Surrey.” Was she the analogue of a modern Park Avenue woman of means, carefully scraping up after her dog: one bag for a cleaner New York, the other for Science with a capital S?

  The pleasure of reading Darwin’s worm book lies not only in recognizing its larger point but also in the charm of detail that Darwin provides about worms themselves. I would rather peruse 300 pages of Darwin on worms than slog through 30 pages of eternal verities explicitly preached by many writers. The worm book is a labor of love and intimate, meticulous detail. In the book’s other major section, Darwin spends 100 pages describing experiments to determine which ends of leaves (and triangular paper cutouts, or abstract “leaves”) worms pull into their burrows first. Here we also find an overt and an underlying theme, in this case leaves and burrows versus the evolution of instinct and intelligence, Darwin’s concern with establishing a usable definition of intelligence, and his discovery (under that definition) that intelligence pervades “lower” animals as well. All great science is a fruitful marriage of detail and generality, exultation and explanation. Both Darwin and his beloved worms left no stone unturned.

  I have argued that Darwin’s last book is a work on two levels—an explicit treatise on worms and the soil and a covert discussion of how to learn about the past by studying the present. But was Darwin consciously concerned with establishing a methodology for historical science, as I have argued, or did he merely stumble into such generality in his last book? I believe that his worm book follows the pattern of all his other works, from first to last: every compendium on minutiae is also a treatise on historical reasoning—and each book elucidates a different principle.

  Darwin’s original illustration for his theory of coral reefs. Top figure: lower solid line, stage 1, a fringing reef (AB) abuts the shore line. Island sinks (level of sea rises) to upper
dotted line, stage 2, barrier reef (A’) separated from sinking island by a lagoon (C).

  Bottom figure: lower solid line, stage 2, barrier reef (copied from upper dotted line of top figure). Island sinks further (below level of sea) to upper dotted line, stage 3, an atoll (A”), enlarged lagoon (C’) marks previous location of sunken island.

  Consider his first book on a specific subject, The Structure and Distribution of Coral-Reefs (1842). In it, he proposed a theory for the formation of atolls, “those singular rings of coral-land which rise abruptly out of the unfathomable ocean,” that won universal acceptance after a century of subsequent debate. He argued that coral reefs should be classified into three categories—fringing reefs that abut an island or continent, barrier reefs separated from island or continent by a lagoon, and atolls, or rings of reefs, with no platform in sight. He linked all three categories with his “subsidence theory,” rendering them as three stages of a single process: the subsidence of an island or continental platform beneath the waves as living coral continues to grow upward. Initially, reefs grow right next to the platform (fringing reefs). As the platform sinks, reefs grow up and outward, leaving a separation between sinking platform and living coral (a barrier reef). Finally the platform sinks entirely, and a ring of coral expresses its former shape (an atoll). Darwin found the forms of modern reefs “inexplicable, excepting on the theory that their rocky bases slowly and successively sank beneath the level of the sea, whilst the corals continued to grow upwards.”

  This book is about coral, but it is also about historical reasoning. Vegetable mold formed fast enough to measure its rate directly; we capture the past by summing effects of small and observable present causes. But what if rates are too slow, or scales too large, to render history by direct observation of present processes? For such cases, we must develop a different method. Since large-scale processes begin at different times and proceed at diverse rates, the varied stages of different examples should exist simultaneously in the present. To establish history in such cases, we must construct a theory that will explain a series of present phenomena as stages of a single historical process. The method is quite general. Darwin used it to explain the formation of coral reefs. We invoke it today to infer the history of stars. Darwin also employed it to establish organic evolution itself. Some species are just beginning to split from their ancestors, others are midway through the process, still others are on the verge of completing it.

  But what if evidence is limited to the static object itself? What if we can neither watch part of its formation nor find several stages of the process that produced it? How can we infer history from a lion? Darwin treated this problem in his treatise on the fertilization of orchids by insects (1862); the book that directly followed the Origin of Species. I have discussed his solution in several essays (1, 4, 11 and The Panda’s Thumb) and will not dwell on it here: we infer history from imperfections that record constraints of descent. The “various contrivances” that orchids use to attract insects and attach pollen to them are the highly altered parts of ordinary flowers, evolved in ancestors for other purposes. Orchids work well enough, but they are jury-rigged to succeed because flowers are not optimally constructed for modification to these altered roles. If God wanted to make insect attractors and pollen stickers from scratch, he would certainly have built differently.

  Thus, we have three principles for increasing adequacy of data: if you must work with a single object, look for imperfections that record historical descent; if several objects are available, try to render them as stages of a single historical process; if processes can be directly observed, sum up their effects through time. One may discuss these principles directly or recognize the “little problems” that Darwin used to exemplify them: orchids, coral reefs, and worms—the middle book, the first, and the last.

  Darwin was not a conscious philosopher. He did not, like Huxley and Lyell, write explicit treatises on methodology. Yet I do not think he was unaware of what he was doing, as he cleverly composed a series of books at two levels, thus expressing his love for nature in the small and his ardent desire to establish both evolution and the principles of historical science. I was musing on this issue as I completed the worm book two weeks ago. Was Darwin really conscious of what he had done as he wrote his last professional lines, or did he proceed intuitively, as men of his genius sometimes do? Then I came to the very last paragraph, and I shook with the joy of insight. Clever old man; he knew full well. In his last words, he looked back to his beginning, compared those worms with his first corals, and completed his life’s work in both the large and the small:

  The plough is one of the most ancient and most valuable of man’s inventions; but long before he existed the land was in fact regularly ploughed, and still continues to be thus ploughed by earthworms. It may be doubted whether there are many other animals which have played so important a part in the history of the world, as have these lowly organized creatures. Some other animals, however, still more lowly organized, namely corals, have done more conspicuous work in having constructed innumerable reefs and islands in the great oceans; but these are almost confined to the tropical zones.

  At the risk of unwarranted ghoulishness, I cannot suppress a final irony. A year after publishing his worm book, Darwin died on April 19, 1882. He wished to be buried in the soil of his adopted village, where he would have made a final and corporeal gift to his beloved worms. But the sentiments (and politicking) of fellow scientists and men of learning secured a guarded place for his body within the well-mortared floor of Westminster Abbey. Ultimately the worms will not be cheated, for there is no permanence in history, even for cathedrals. But ideas and methods have all the immortality of reason itself. Darwin has been gone for a century, yet he is with us whenever we choose to think about time.

  10 | A Hearing for Vavilov

  IN 1936, TROFIM D. LYSENKO, struggling to reform Russian agricultural science on discredited Lamarckian principles, wrote: “I am not fond of controversy in matters concerning theory. I am an ardent controversialist only when I see that in order to carry out certain practical tasks I must remove the obstacles that stand in the way of my scientific activities.”

  As his practical task, Lysenko set out to “alter the nature of plants in the direction we desire by suitable training.” He argued that previous failure to produce rapid and heritable improvements in important crop plants must be laid to the bankrupt ideology of bourgeois science, with its emphasis on sterile academic theory and its belief in Mendelian genes, which do not respond directly to the prodding of breeders but change only by accidental and random mutation. The criterion of a more adequate science must be success in improved breeding.

  “The better we understand the laws of development of plant and animal forms,” he wrote, “the more easily and quickly will we be able to create the forms we need in accordance with our wishes and plans.” What “laws of development” could be more promising than the Lamarckian claim that altered environments can directly induce heritable changes in desired directions? If only Nature worked this way! But she does not, and all Lysenko’s falsified data and vicious polemics budged her not one inch.

  If Lysenko’s “obstacles” had been disembodied ideas alone, the history of Russian genetics might have been spared some of its particular tragedy. But ideas emanate from people, and the obstacles designated for removal were necessarily human. Nikolai Ivanovich Vavilov, Russia’s leading Mendelian geneticist and director of the All-Union Lenin Academy of Agricultural Sciences centered in Leningrad, served as a focal point for Lysenko’s attacks in 1936. Lysenko castigated Vavilov for his general Mendelian views, but any geneticist might have served equally well for such generalized target practice. Lysenko singled out Vavilov for a more specific and personal theory (and the subject of this column)—the so-called law of homologous series in variation.

  Twelve years later, following the devastation of war, Lysenko had triumphed. His infamous address, “The Situation in Biological Science,”
read at the 1948 session of the Lenin Academy of Agricultural Sciences, contains as the first statement of its summary what may well be the most chilling passage in all the literature of twentieth-century science.

  The question is asked in one of the notes handed to me, “What is the attitude of the Central Committee of the Party to my report?” I answer: The Central Committee of the Party has examined my report and approved it [Stormy applause. Ovation. All rise].

  Following another ten pages of rhetoric and invective, Lysenko concludes: “Glory to the great friend and protagonist of science, our leader and teacher, Comrade Stalin! [All rise. Prolonged applause.]”

  Nikolai Vavilov was unable to attend the 1948 meeting. He had been arrested in August 1940 while on a collecting expedition in the Ukraine. In July 1941, he was sentenced to death for agricultural sabotage, spying for England, maintaining links with émigrés, and belonging to a rightist organization. The sentence was commuted to ten years imprisonment, and Vavilov was moved to the inner prison of the NKVD in Moscow. In October, he was evacuated to the Saratov prison where he spent several months in an underground death cell, suffering from malnutrition. He died, still a prisoner, in January 1943.

  What is Vavilov’s “law of homologous series in variation,” and how did it provide Lysenko with rhetorical leverage? Vavilov published this law, the guiding principle for much of his practical work in agricultural genetics, in 1920 and revised it in 1935. It was printed in English in the prestigious Journal of Genetics in 1922 (vol. 12, pp. 48–89).

  Vavilov was perhaps the world’s leading expert on the biogeography of wheat and other cereals. He traveled throughout the world (thereby leaving himself vulnerable to trumped-up charges of espionage), collecting varieties of plants from their natural habitats and establishing the world’s largest “bank” of genetic variation within major agricultural species. As he collected natural races of wheat, barley, oats, and millet over a large range of environments and places, he noticed that strikingly similar series of varieties could be found within the different species of a genus and often within species of related genera as well.