Unlike Darwin, Geoffroy or Spallanzani, there was never any doubt about William Bateson’s career. Academia beckoned from the start: from Rugby School he went to St John’s College, Cambridge (where his father was Master) and took a double first in natural sciences. After a spell at Johns Hopkins University in Baltimore, he returned to Cambridge, but left again in 1910 to become the first director of the John Innes Institute in Norwich, a pre-eminent centre for the new science of genetics.
Early on, Bateson acquired the self-confidence of one who is convinced that he is in the right and everyone else is wrong. ‘He formed definite opinions on a number of subjects, from the Sistine Madonna and compulsory Greek to nationalism and natural selection,’ wrote the biologist J. B. S. Haldane in a review of Bateson’s posthumously collected essays and addresses.1 By the same token, Bateson was not always ready to accommodate the views of others, something his peers found a ‘source of regret’, according to Bateson’s American colleague Thomas Hunt Morgan. 2
Like many Victorian zoologists in the decades immediately after the Origin, Bateson was interested in tracing the early evolution of the vertebrates – the backboned animals, including ourselves – by comparison with our invertebrate relatives such as sea squirts and lancelets. Bateson was particularly interested in acorn worms, obscure residents of muddy beaches and ocean bottoms. These creatures, hardly known outside a small coterie of zoologists, are rather unpleasant-looking, flaccid creatures that range from a couple of centimetres to a hundred times that length. Despite their unprepossessing exterior, they have long fascinated zoologists because they seem to bridge a gap between two very different animal groups. They have a series of gill slits reminiscent of those of fishes, but as microscopic larvae floating around in the sea, before adopting the settled life of the adult, they look very like the larvae of echinoderms, the group of animals that includes starfishes and sea cucumbers.
In the 1880s, when Bateson began his work in this field, the origin of vertebrates was a hot topic, and the study of transitional creatures such as acorn worms was seen as a way to find out about a period of evolution now long vanished and yet crucial to the understanding of our own place in nature. Of particular interest to Bateson – as it remains to scientists now – was understanding the evolution of those features that make vertebrates distinctive, such as the head and the serial arrangement of muscles which allows rapid swimming. The superficially fish-like lancelet, being most closely similar to vertebrates, was the focus of special attention. Lancelets have segmental muscles arranged in a nose-to-tail series, just like fishes. Tunicates, in contrast, generally spend their adult lives rooted to the same spot, filtering particles of food from sea water strained through their gill slits. Tunicate larvae are more mobile, each propelled by a tail supported by a notochord, which are both lost when the larva settles and metamorphoses into an adult.
Several scientists looked at these facts and wove them into elaborate evolutionary scenarios. Between the 1890s and the 1920s, an English marine biologist called Walter Garstang (1868–1949), who in his regular occupation had the more mundane task of monitoring fisheries, suggested that long ago some tunicates forwent their adult stage and remained as tadpole-like larvae throughout life, gradually evolving into vertebrates. This idea was elaborated by others, notably the English tunicate specialist N. J. Berrill (1903-96), as late as the 1950s. According to Berrill, these evolving animals acquired serial muscle blocks as adaptations for efficient swimming, especially against the current. Life originated in the sea, but it was supposed that vertebrates had evolved in freshwater. On the basis of this assumption (now known to be flawed), Berrill argued that the immediate ancestors of vertebrates must have swum against the current to find fresh water – where they would have needed to evolve kidneys as a way of excreting the water absorbed by their relatively salt-rich tissues. Kidneys are organs unique to vertebrates, not found in lancelets. Skulls and brains came later, as adaptations for a new role as predators – and behold, vertebrates appeared. Lancelets, however, remained at sea – evolutionary laggards unsuited for energetic migration to freshwater. This is less a scientific hypothesis than a latter-day fairy story that brings in natural selection where needed, as a magical ingredient which transforms one animal into another, as surely as a magic spell can turn a frog into a prince, and leave a toad, or a lancelet, untouched.
In 1909, the Linnean Society staged a debate on the origin of vertebrates to commemorate the golden jubilee of the publication of the Origin of Species – which had first emerged in the papers of Wallace and Darwin, read to the Society in 1858.3 On one side was a physiologist named Walter Gaskell and his colleagues, who held that vertebrates were, in essence, highly modified crabs. Opposing them were arrayed the great zoological names of the day, who held that lancelets were the closest relatives of vertebrates (as is the consensus today). One of the participants, Thomas Stebbing – a cleric, and perhaps a neutral voice – remarked that:
When we return home and our friends gleefully enquire, ‘What then has been decided as to the Origin of Vertebrates?’, so far we seem to have no reply ready, except that the disputants agreed on one single point, namely, that their opponents were all in the wrong.4
The debate at the Linnean was the last echo of what, by then, was a long-outmoded style of evolutionary speculation.
By that time, Bateson had abandoned the quest for the origin of vertebrates as so much vacuous storytelling, but the seeds of his dissatisfaction were sown early. During visits to America in 1884 and 1885 to collect specimens of acorn worms, he had met William Keith Brooks (1848-1908) of Johns Hopkins University, a passionate evolutionist – but one who also liked to ask the kind of difficult questions about evolution that Bateson’s colleagues at home seemed content to avoid. Brooks impressed on Bateson the importance of the ‘species question’ in biology. Darwin says that new species evolve from old ones – but how? Is natural selection the answer? And what, anyway, is a species? What is the nature of the variation on which selection acts? Brooks was interested in the possibility that species did not transmute gradually, but all of a sudden, in a series of jumps. Bateson was gripped by this idea of evolution through sudden mutation rather than the ‘insensible gradations’ proposed by Darwin, and following his meeting with Brooks his career took a new turn.
By 1894, Bateson’s views on the search for the origin of vertebrates would have made a fine review of the Linnean debate of 1909, for all that they came fifteen years ahead of the event. ‘Were we all agreed in our assumptions and as to the canons of interpretation, there might be some excuse, but we are not agreed,’ he writes, in the introduction to Materials for the Study of Variation:
Out of the same facts of anatomy and development men of equal ability and repute have brought the most opposite conclusions. To take for instance the question of the ancestry of the Chordata, the problem on which I was myself engaged, even if we neglect fanciful suggestions, there remain two wholly incompatible views as to the lines of Vertebrate descent, each well supported and upheld by many. From the same facts opposite conclusions are drawn.5
Bateson then drew a line in the sand, ending his career as an evolutionary biologist: ‘Facts of the same kind will take us no further. The issue turns not on the facts but on the assumptions. Surely we can do better than this. Need we waste more effort in these vain and sophistical disputes?’6 With these lines, Bateson went straight to the heart of the problem: if scientists were to make any progress towards understanding evolution, they had to try to make sense of the nature and sources of the variation on which selection acts. Natural selection may or may not be fine as a mechanism, but with no comprehension of the sources of variation we would come no nearer to understanding the nature of that which creates and maintains an organism – what we now know as the genome.
It is important to remember that Bateson wrote Materials before anyone had any clear idea of the mechanisms of inheritance, still less of genes or genomes. He knew that he had to start at the beginning, with a comprehensive catalogue of all instances of natural variation he could find, from which he hoped that any underlying laws of variation might emerge. To this end he scoured libraries, searched in museums at home and abroad, and assembled his own collections from as far away as Central Asia and Siberia. The resulting work – Materials for the Study of Variation – is not so much a book as an encyclopedia. It is a hefty compendium of no fewer than 886 reported deviations from the norm: variations in the numbers of teeth, horns, scales, bristles, body segments, sex organs, legs, antennae, digits, coloration and tentacles in a vast range of organisms, from people to pythons, shrimps to sloths.
Bateson was especially fascinated by the same patterns that had intrigued Geoffroy and Goethe – in particular in the repetition of parts, whether leaves on a stem, teeth in a jaw, or segments in the body of an insect, or a fish – and how individual elements in a series of repeated elements may be specialized. He used the term ‘meristic’ to refer to this kind of serial repetition. An example might be the legs of a segmented animal such as a crustacean – a lobster or shrimp. Most segments of the body of a crustacean bear paired limbs. In one sense, the limbs are ‘the same’ in that they are all repeated units which have much in common as regards their position (paired, attached to body segments) and structure (division into a set number of articulating parts). In another sense, each limb is specialized according to its function. Limbs associated with the head may have been turned into claws for grasping prey, or jaws for processing the prey, once caught; limbs farther back are used as walking legs; the limbs at the end are paddles, for swimming.
But Bateson also discussed disturbances in such ordered series. He described as an ‘extraordinary Discontinuity of Variation’7 those cases in which one element in a repeated series may be supplanted by that of another with a brazen neatness in its substitution. The vertebrae in the human spinal column provided several good examples. They are not identical repeated units, but the column as a whole is clearly divisible into several regions, the vertebrae in each region having a distinctive form. Like the legs of lobsters, the vertebrae in the human spine all clearly belong to a repeated series, but are also specialized according to their region. The skull is supported by the highly distinctive atlas and axis vertebrae, at the top of a series of cervical (neck) vertebrae. These vertebrae do not normally bear ribs, unlike the vertebrae in the next, thoracic region. Below the thoracic vertebrae come the stout, non-rib-bearing lumbar vertebrae of the lower back; the sacral vertebrae, fused into a single bone, the sacrum, to which the pelvis and lower limbs are attached; and finally the vertebrae forming the coccyx, the tiny vestige of a tail. Bateson catalogued instances, gleaned from anatomy, in which the lowermost cervical vertebrae tend, unusually, to bear ribs so that they resemble the uppermost vertebrae in the next (thoracic) region. In other cases, the lowest thoracic vertebra is uncharacteristically free of ribs and so resembles, instead, the uppermost lumbar vertebra – and vice versa.
More strikingly, Bateson recorded instances in insects and other arthropods (jointed-limbed animals) in which one kind of appendage appears in place of another. His case no. 76, for example, is a specimen of the bee Bombus variabilis ‘taken beside the hedge of a park in Munich, having the left antenna partially developed as a foot’.8 Case no. 78, in contrast, is a moth, Zygaena filipendulae, which possesses a ‘supernumerary wing arising in such a position as to suggest it replaced a leg’.9 Bateson coined a word, homeosis, for the situation in which one kind of appendage or member in a series (such as an insect leg, or a human thoracic vertebra) appears in the place usually occupied by another (such as an insect antenna, or a human cervical vertebra). He could hardly have predicted the importance that homeosis would assume, a century later, in our understanding of the origin of form.
Bateson’s fascination with meristic variation – Materials is dominated by it – speaks to a deeper concern, and suggests a fundamental difference in approach from Darwin’s. Whereas Darwin travelled the world and saw abundant diversity crying out for explanation, Bateson catalogued the minutiae of museum collections and saw not diversity, but unity. Parts may vary in a series, but we would not recognize the series at all without some kind of underlying similarity.
Conventional wisdom has it that 1859, the year in which the Origin of Species was first published, is a kind of biological Year Zero, such that discussion of work before that date becomes more of historical than of scientific interest. If this was the view in 1894, it is not reflected in Materials: Bateson’s preoccupations with unity and diversity clearly owe much to pre-Darwinian ideas – ideas which, in the absence (to him) of any credible Darwinian consensus, were still vital and relevant. Nature-philosophy was, of course, built on the same preoccupations: Goethe, for example, had written on such things as the underlying similarities of the organs of plants, suggesting that structures as diverse as petals, sepals and stamens could be thought of as modified leaves. Echoes of the debate between Geoffroy and Cuvier can clearly be heard in Bateson’s discussions of monstrosity, of vertebrae and of the structure of insects.
However, Materials plumbs even deeper seams of thought than these, showing just how fundamental and ambitious a revision of biological thought Bateson had intended to undertake. For Materials, in its structure and intention, is really a kind of medieval bestiary, a genre harking back to the very first stirrings of modern scientific thought. Bestiaries were illustrated compendia of medieval zoological knowledge in which everyday animals were depicted alongside beasts from mythology or folklore, as if the two kinds were equally real. In an age when the world outside Europe was just beginning to be explored, bestiaries were also repositories of travellers’ lore and, as such, were enthralling entertainments. Shakespeare’s Othello tells how Desdemona’s love for him was kindled by stories of his own adventures in faraway places, in which he met an assortment of real and mythical beings:
of the Cannibals that each other eat,
The Anthropophagi, and men whose heads
Do grow beneath their shoulders.
Parallelling the bestiaries were catalogues of human and animal monstrosities – deformities of birth. Galleries of monsters started out as freak shows, doubling as moral fables to terrify sinners into contrition. Stripped of their mystical, folkloric and moral content, these works became the first medical textbooks, as physicians began to see in monsters a way of testing the pattern of nature. If monsters were offences against nature that pointed up the normal range of variation, then monstrosities might be catalogued in a systematic way, so that the sources of this variation might be better understood.
One of the first to catalogue monstrosity in an objective fashion was Ambroise Paré (1493-1541), a contemporary of Paracelsus and surgeon to the French monarchy, whose Monsters and Marvels was published posthumously in 1573. Paré classified monsters according to their supposed causes, which ranged from the theological (consequences of the wrath of God) and fanciful (the result of having been cursed by a beggar) to the practical (injury or illness sustained to the mother while pregnant) and even prescient (inheritance). In 1620, the English philosopher Francis Bacon (1561–1626) recommended to his fellow examiners of nature that they make compilations of natural monstrosity as a way of understanding the boundaries of what is normal. The occasional dog or sheep born with an extra pair of spindly legs, or an extra head, serves to emphasize that farm animals should be expected to have four legs and but one head.
As historical phenomena, bestiaries and galleries of monsters can be seen as expressions of ignorance as much as early explorations of nature’s plan, and they had all but disappeared by the end of the eighteenth century. The fact that Materials is composed along the lines of a bestiary, and has precisely the purpose laid down by Francis Bacon – that is, to attempt an understanding of the sources of variation by learning the demarcation lines between the expected and the monstrous – can be seen as Bateson’s way of telling us that, despite Darwin, and indeed despite everything, our apprehension of the fundamental laws of form, generation and heredity remained as rudimentary in 1894 as it was in the days of alchemy. Bateson’s moths with legs growing out of their heads are every bit as freakish and unexplained as Othello’s men whose heads grow beneath their shoulders.
Bateson’s emphasis on the importance of the monstrous, of gross deviation from the norm, put him at odds with the Darwinian view that change comes in very small increments. Where Darwin’s mission was to explain diversity through variation whose existence was a given, Bateson sought underlying rules that could explain the disposition, and thus the existence, of the variation on which natural selection depended.
In addition, Bateson was dissatisfied with Darwin’s thesis of gradual change, because the evidence of his own eyes told a different story. The variations recorded by Bateson in Materials tended not to be subtle matters of degree, but prominent changes in kind. There were creatures with too few digits, or too many. But there were none with fractions of digits, caught in some act of gradual atrophy or acquisition. Qualitative, disjunct variation, thought Bateson, was hardly what Darwin was talking about when he spoke of gradual evolutionary change that took geological ages to complete. Plainly, there was more to variation than Darwin suspected: it could be that the transition between species could be by way of this discontinuous variation, rather than the gradual, gentle process envisaged by Darwin.
Writing with the passion of Goethe, Bateson asserted that discontinuous variation could not be the work of selection:
the existence of sudden and discontinuous Variation, the existence, that is to say, of new forms having from their first beginning more or less of the kind of perfection that we associate with normality, is a fact that disposes, once and for all, of the attempt to interpret all perfection and definiteness of form as the work of Selection. The study of Variation leads us into the presence of whole classes of phenomena that are plainly incapable of such interpretation. The existence of Discontinuity in Variation is therefore a final proof that the accepted hypothesis is inadequate.10
The idea of ‘perfection’ owes much to nature-philosophy – and nothing to Darwin. If there is any doubt about Bateson’s views, he lays them to rest later on: ‘it is quite certain that the distinctness and Discontinuity of many characters is in some unknown way a part of their nature, and is not directly dependent on Natural Selection at all’.11 The phrase ‘some unknown way’ suggests that Bateson, like Darwin, could not explain the nature and sources of this variation. He was, like William Harvey centuries earlier, disarmingly candid about his ignorance. His critics – and there were many – could only ever take this admission as an accusation that there was something missing at the heart of Darwinism. Bateson attracted trenchant criticism from Darwin’s direct intellectual descendants.
In an age in which the presentation of science was becoming increasingly quantitative and based on experiment, the self-styled inheritors of the Darwinian flame were embarrassed by Darwin’s anecdotal style and sought to put his ideas on a soundly mathematical footing. If evolution were a matter of minute steps, then this minuteness had to be measurable, and so the ‘bio-metrical’ school came into being. Prominent biometricians included Galton, the statistician Karl Pearson (1857-1936), Walter Weldon (1860-1906) and Edward Poulton (1856-1943).
The biometricians reasoned that if evolution happened in very small steps, the overall effect would be one of gradual change, and that the nature of variation of a trait in a population would be ‘continuous’. In other words, the discrete but slight variation between individuals could be expressed as a continuum, so that both extremes can be connected by a series of intermediates. For example, the human race as a whole shows a vast range of skin pigmentation, from almost white to deeply black. But individuals can be found with intermediate skin tones, and if you found enough individuals and lined them up in order of pigmentation, the transition from white to black would be so smooth as to appear continuous. Any idea that variation might be discontinuous, or that discontinuous variation of the monstrous kind suggested by Bateson had any part to play in evolution, was anathema to the biometricians. Bateson’s almost histrionic self-importance only made matters worse: Poulton castigated Bateson for his ‘narrowness, dogmatism, prejudice, and contemptuous deprecation of research about which he was regrettably ill-informed, not to mention [his] exaggerated estimation of the importance of his own work’.12 To some, Bateson’s stance against Darwin looked like betrayal: that Bateson and Weldon had once been close colleagues made the battle all the more personal.
The wounds opened by Materials would not close for more than four decades, and in a sense they have not fully healed even today. Modern commentators recognize Bateson for his pioneering work in genetics, while quietly ignoring Materials as if it were some kind of prehistoric aberration rendered obsolete once its author had invented the science of genetics just a few years later. Bateson’s arrogance and his overwrought literary style have clearly prevented his contribution from being properly recognized, but there is something else at work – something of the nature of revisionism. From the perspective of a modern Darwinian, the fact that a pioneer of genetics as important as Bateson held such passionately anti-Darwinian views is embarrassing, and so Materials is politely but firmly ignored.
Once this prejudice has been taken into account, and Materials reappraised with the perspective that only a century of distance can offer, Bateson’s bestiary looks less like the irrelevant ravings of a rebel whose cause had yet to be found than a pivot around which biology turns. Between a single set of covers is a book that looks back to the very dawn of our efforts to understand the origin of form, and at the same time looks forward to genetics and the quest for the genome, whose rewards we have only now begun to enjoy.