Darwin had obviously split ways with Lamarck’s evolutionary ideas. Giraffes hadn’t arisen from straining antelopes needing skeletal neck-braces. They had emerged—loosely speaking—because an ancestral antelope had produced a long-necked variant that had been progressively selected by a natural force, such as a famine. But Darwin kept returning to the mechanism of heredity: What had made the long-necked antelope emerge in the first place?
Darwin tried to envision a theory of heredity that would be compatible with evolution. But here his crucial intellectual shortcoming came to the fore: he was not a particularly gifted experimentalist. Mendel, as we shall see, was an instinctual gardener—a breeder of plants, a counter of seeds, an isolator of traits; Darwin was a garden digger—a classifier of plants, an organizer of specimens, a taxonomist. Mendel’s gift was experimentation—the manipulation of organisms, cross-fertilization of carefully selected sub-breeds, the testing of hypotheses. Darwin’s gift was natural history—the reconstruction of history by observing nature. Mendel, the monk, was an isolator; Darwin, the parson, a synthesizer.
But observing nature, it turned out, was very different from experimenting with nature. Nothing about the natural world, at first glance, suggests the existence of a gene; indeed, you have to perform rather bizarre experimental contortions to uncover the idea of discrete particles of inheritance. Unable to arrive at a theory of heredity via experimental means, Darwin was forced to conjure one up from purely theoretical grounds. He struggled with the concept for nearly two years, driving himself to the brink of a mental breakdown, before he thought he had stumbled on an adequate theory. Darwin imagined that the cells of all organisms produce minute particles containing hereditary information—gemmules, he called them. These gemmules circulate in the parent’s body. When an animal or plant reaches its reproductive age, the information in the gemmules is transmitted to germ cells (sperm and egg). Thus, the information about a body’s “state” is transmitted from parents to offspring during conception. As with Pythagoras, in Darwin’s model, every organism carried information to build organs and structures in miniaturized form—except in Darwin’s case, the information was decentralized. An organism was built by parliamentary ballot. Gemmules secreted by the hand carried the instructions to manufacture a new hand; gemmules dispersed by the ear transmitted the code to build a new ear.
How were these gemmular instructions from a father and a mother applied to a developing fetus? Here, Darwin reverted to an old idea: the instructions from the male and female simply met in the embryo and blended together like paints or colors. This notion—blending inheritance—was already familiar to most biologists: it was a restatement of Aristotle’s theory of mixing between male and female characters. Darwin had, it seemed, achieved yet another marvelous synthesis between opposing poles of biology. He had melded the Pythagorean homunculus (gemmules) with the Aristotelian notion of message and mixture (blending) into a new theory of heredity.
Darwin dubbed his theory pangenesis—“genesis from everything” (since all organs contributed gemmules). In 1867, nearly a decade after the publication of Origin, he began to complete a new manuscript, The Variation of Animals and Plants Under Domestication, in which he would fully explicate this view of inheritance. “It is a rash and crude hypothesis,” Darwin confessed, “but it has been a considerable relief to my mind.” He wrote to his friend Asa Gray, “Pangenesis will be called a mad dream, but at the bottom of my own mind, I think it contains a great truth.”
Darwin’s “considerable relief” could not have been long-lived; he would soon be awoken from his “mad dream.” That summer, while Variation was being compiled into its book form, a review of his earlier book, Origin, appeared in the North British Review. Buried in the text of that review was the most powerful argument against pangenesis that Darwin would encounter in his lifetime.
The author of the review was an unlikely critic of Darwin’s work: a mathematician-engineer and inventor from Edinburgh named Fleeming Jenkin, who had rarely written about biology. Brilliant and abrasive, Jenkin had diverse interests in linguistics, electronics, mechanics, arithmetic, physics, chemistry, and economics. He read widely and profusely—Dickens, Dumas, Austen, Eliot, Newton, Malthus, Lamarck. Having chanced upon Darwin’s book, Jenkin read it thoroughly, worked swiftly through the implications, and immediately found a fatal flaw in the argument.
Jenkin’s central problem with Darwin was this: if hereditary traits kept “blending” with each other in every generation, then what would keep any variation from being diluted out immediately by interbreeding? “The [variant] will be swamped by the numbers,” Jenkin wrote, “and after a few generations its peculiarity will be obliterated.” As an example—colored deeply by the casual racism of his era—Jenkin concocted a story: “Suppose a white man to have been wrecked on an island inhabited by negroes. . . . Our shipwrecked hero would probably become king; he would kill a great many blacks in the struggle for existence; he would have a great many wives and children.”
But if genes blended with each other, then Jenkin’s “white man” was fundamentally doomed—at least in a genetic sense. His children—from black wives—would presumably inherit half his genetic essence. His grandchildren would inherit a quarter; his great-grandchildren, an eighth; his great-great-grandchildren, one-sixteenth, and so forth—until his genetic essence had been diluted, in just a few generations, into complete oblivion. Even if “white genes” were the most superior—the “fittest,” to use Darwin’s terminology—nothing would protect them from the inevitable decay caused by blending. In the end, the lone white king of the island would vanish from its genetic history—even though he had fathered more children than any other man of his generation, and even though his genes were best suited for survival.
The particular details of Jenkin’s story were ugly—perhaps deliberately so—but its conceptual point was clear. If heredity had no means of maintaining variance—of “fixing” the altered trait—then all alterations in characters would eventually vanish into colorless oblivion by virtue of blending. Freaks would always remain freaks—unless they could guarantee the passage of their traits to the next generation. Prospero could safely afford to create a single Caliban on an isolated island and let him roam at large. Blending inheritance would function as his natural genetic prison: even if he mated—precisely when he mated—his hereditary features would instantly vanish into a sea of normalcy. Blending was the same as infinite dilution, and no evolutionary information could be maintained in the face of such dilution. When a painter begins to paint, dipping the brush occasionally to dilute the pigment, the water might initially turn blue, or yellow. But as more and more paints are diluted into the water, it inevitably turns to murky gray. Add more colored paint, and the water remains just as intolerably gray. If the same principle applied to animals and inheritance, then what force could possibly conserve any distinguishing feature of any variant organism? Why, Jenkin might ask, weren’t all Darwin’s finches gradually turning gray?I
Darwin was deeply struck by Jenkin’s reasoning. “Fleeming Jenkins [sic] has given me much trouble,” he wrote, “but has been of more use to me than any other Essay or Review.” There was no denying Jenkin’s inescapable logic: to salvage Darwin’s theory of evolution, he needed a congruent theory of heredity.
But what features of heredity might solve Darwin’s problem? For Darwinian evolution to work, the mechanism of inheritance had to possess an intrinsic capacity to conserve information without becoming diluted or dispersed. Blending would not work. There had to be atoms of information—discrete, insoluble, indelible particles—moving from parent to child.
Was there any proof of such constancy in inheritance? Had Darwin looked carefully through the books in his voluminous library, he might have found a reference to an obscure paper by a little-known botanist from Brno. Unassumingly entitled “Experiments in Plant Hybridization” and published in a scarcely read journal in 1866, the paper was written in dense German and packed wi
th the kind of mathematical tables that Darwin particularly despised. Even so, Darwin came tantalizingly close to reading it: in the early 1870s, poring through a book on plant hybrids, he made extensive handwritten notes on pages 50, 51, 53, and 54—but mysteriously skipped page 52, where the Brno paper on pea hybrids was discussed in detail.
If Darwin had actually read it—particularly as he was writing Variation and formulating pangenesis—this study might have provided the final critical insight to understand his own theory of evolution. He would have been fascinated by its implications, moved by the tenderness of its labor, and struck by its strange explanatory power. Darwin’s incisive intellect would quickly have grasped its implications for the understanding of evolution. He may also have been pleased to note that the paper had been authored by another cleric who, in another epic journey from theology to biology, had also drifted off the edge of a map—an Augustine monk named Gregor Johann Mendel.
* * *
I. Geographic isolation might have solved some of the “grey finch” problem—by restricting interbreeding between particular variants. But this would still be unable to explain why all finches in a single island did not gradually collapse to have identical characteristics.
“Flowers He Loved”
We want only to disclose the [nature of] matter and its force. Metaphysics is not our interest.
—The manifesto of the Brünn Natural Science Society, where Mendel’s paper was first read in 1865
The whole organic world is the result of innumerable different combinations and permutations of relatively few factors. . . . These factors are the units which the science of heredity has to investigate. Just as physics and chemistry go back to molecules and atoms, the biological sciences have to penetrate these units in order to explain . . . the phenomena of the living world.
—Hugo de Vries
As Darwin was beginning to write his opus on evolution in the spring of 1856, Gregor Mendel decided to return to Vienna to retake the teacher’s exam that he had failed in 1850. He felt more confident this time. Mendel had spent two years studying physics, chemistry, geology, botany, and zoology at the university in Vienna. In 1853, he had returned to the monastery and started work as a substitute teacher at the Brno Modern School. The monks who ran the school were very particular about tests and qualifications, and it was time to try the certifying exam again. Mendel applied to take the test.
Unfortunately, this second attempt was also a disaster. Mendel was ill, most likely from anxiety. He arrived in Vienna with a sore head and a foul temper—and quarreled with the botany examiner on the first day of the three-day test. The topic of disagreement is unknown, but likely concerned species formation, variation, and heredity. Mendel did not finish the exam. He returned to Brno reconciled to his destiny as a substitute teacher. He never attempted to obtain certification again.
Late that summer, still bruising from his failed exam, Mendel planted a crop of peas. It wasn’t his first crop. He had been breeding peas inside the glass hothouse for about three years. He had collected thirty-four strains from the neighboring farms and bred them to select the strains that bred “true”—that is, every pea plant produced exactly identical offspring, with the same flower color or the same seed texture. These plants “remained constant without exception,” he wrote. Like always begat like. He had collected the founding material for his experiment.
The true-bred pea plants, he noted, possessed distinct traits that were hereditary and variant. Bred to themselves, tall-stemmed plants generated only tall plants; short plants only dwarf ones. Some strains produced only smooth seeds, while others produced only angular, wrinkled seeds. The unripe pods were either green or vividly yellow, the ripe pods either loose or tight. He listed the seven such true-breeding traits:
1. the texture of the seed (smooth versus wrinkled)
2. the color of seeds (yellow versus green)
3. the color of the flower (white versus violet)
4. the position of the flower (at the tip of the plant versus the branches)
5. the color of the pea pod (green versus yellow)
6. the shape of the pea pod (smooth versus crumpled)
7. the height of the plant (tall versus short)
Every trait, Mendel noted, came in at least two different variants. They were like two alternative spellings of the same word, or two colors of the same jacket (Mendel experimented with only two variants of the same trait, although, in nature, there might be multiple ones, such as white-, purple-, mauve-, and yellow-flowering plants). Biologists would later term these variants alleles, from the Greek word allos—loosely referring to two different subtypes of the same general kind. Purple and white were two alleles of the same trait: flower color. Long and short were two alleles of another characteristic—height.
The purebred plants were only a starting point for his experiment. To reveal the nature of heredity, Mendel knew that he needed to breed hybrids; only a “bastard” (a word commonly used by German botanists to describe experimental hybrids) could reveal the nature of purity. Contrary to later belief, he was acutely aware of the far-reaching implication of his study: his question was crucial to “the history of the evolution of organic forms,” he wrote. In two years, astonishingly, Mendel had produced a set of reagents that would allow him to interrogate some of the most important features of heredity. Put simply, Mendel’s question was this: If he crossed a tall plant with a short one, would there be a plant of intermediate size? Would the two alleles—shortness and tallness—blend?
The production of hybrids was tedious work. Peas typically self-fertilize. The anther and the stamen mature inside the flower’s clasplike keel, and the pollen is dusted directly from a flower’s anther to its own stamen. Cross-fertilization was another matter altogether. To make hybrids, Mendel had to first neuter each flower by snipping off the anthers—emasculating it—and then transfer the orange blush of pollen from one flower to another. He worked alone, stooping with a paintbrush and forceps to snip and dust the flowers. He hung his outdoor hat on a harp, so that every visit to the garden was marked by the sound of a single, crystalline note. This was his only music.
It’s hard to know how much the other monks in the abbey knew about Mendel’s experiments, or how much they cared. In the early 1850s, Mendel had tried a more audacious variation of this experiment, starting with white and gray field mice. He had bred mice in his room—mostly undercover—to try to produce mice hybrids. But the abbot, although generally tolerant of Mendel’s whims, had intervened: a monk coaxing mice to mate to understand heredity was a little too risqué, even for the Augustinians. Mendel had switched to plants and moved the experiments to the hothouse outside. The abbot had acquiesced. He drew the line at mice, but didn’t mind giving peas a chance.
By the late summer of 1857, the first hybrid peas had bloomed in the abbey garden, in a riot of purple and white. Mendel noted the colors of the flowers, and when the vines had hung their pods, he slit open the shells to examine the seeds. He set up new hybrid crosses—tall with short; yellow with green; wrinkled with smooth. In yet another flash of inspiration, he crossed some hybrids to each other, making hybrids of hybrids. The experiments went on in this manner for eight years. The plantings had, by then, expanded from the hothouse to a plot of land by the abbey—a twenty-foot-by-hundred-foot rectangle of loam that bordered the refectory, visible from his room. When the wind blew the shades of his window open, it was as if the entire room turned into a giant microscope. Mendel’s notebook was filled with tables and scribblings, with data from thousands of crosses. His thumbs were getting numb from the shelling.
“How small a thought it takes to fill someone’s whole life,” the philosopher Ludwig Wittgenstein wrote. Indeed, at first glance, Mendel’s life seemed to be filled with the smallest thoughts. Sow, pollinate, bloom, pluck, shell, count, repeat. The process was excruciatingly dull—but small thoughts, Mendel knew, often bloomed into large principles. If the powerful scientific revolution that had
swept through Europe in the eighteenth century had one legacy, it was this: the laws that ran through nature were uniform and pervasive. The force that drove Newton’s apple from the branch to his head was the same force that guided planets along their celestial orbits. If heredity too had a universal natural law, then it was likely influencing the genesis of peas as much as the genesis of humans. Mendel’s garden plot may have been small—but he did not confuse its size with that of his scientific ambition.
“The experiments progress slowly,” Mendel wrote. “At first a certain amount of patience was needed, but I soon found that matters went better when I was conducting several experiments simultaneously.” With multiple crosses in parallel, the production of data accelerated. Gradually, he began to discern patterns in the data—unanticipated constancies, conserved ratios, numerical rhythms. He had tapped, at last, into heredity’s inner logic.
The first pattern was easy to perceive. In the first-generation hybrids, the individual heritable traits—tallness and shortness, or green and yellow seeds—did not blend at all. A tall plant crossed with a dwarf inevitably produced only tall plants. Round-seeded peas crossed with wrinkled seeds produced only round peas. All seven of the traits followed this pattern. “The hybrid character” was not intermediate but “resembled one of the parental forms,” he wrote. Mendel termed these overriding traits dominant, while the traits that had disappeared were termed recessive.