Gregory Radick, 2011. “Physics in the Galtonian Sciences of Heredity.” Studies in History and Philosophy of Biological and Biomedical Sciences 42: 129-138.

Abstract

Physics matters less than we once thought to the making of Mendel. But it matters more than we tend to recognize to the making of Mendelism. This paper charts the variety of ways in which diverse kinds of physics impinged upon the Galtonian tradition which formed Mendelism’s matrix. The work of three Galtonians in particular is considered: Francis Galton himself, W. F. R. Weldon and William Bateson. One aim is to suggest that tracking influence from physics can bring into focus important but now little-remembered flexibilities in the Galtonian tradition. Another is to show by example why generalizations about what happens when ‘physics’ meets ‘biology’ require caution. Even for a single research tradition in Britain in the decades around 1900, these categories were large, containing multitudes.

Keywords

  • Physics
  • Biology
  • Francis Galton
  • W. F. R. Weldon
  • William Bateson
  • Mendelism

1. A Mendelian prelude

No general history of biology is more enterprising in drawing out biology’s dependencies on physics than David Depew and Bruce Weber’s Darwinism evolving(1995). Their main concern is with the theory of natural selection, which they see as changing with broader shifts in ‘systems dynamics’ from Newtonianism onward. But they also deal at length with the sciences of heredity, including that ever-appealing hero, Gregor Mendel. For them, Mendel is the man who quantized heredity, bequeathing to the twentieth century its notion of hereditary factors. Reconceived as material particles and regarded from the population level, genes–as the factors came to be known–enabled Darwinian biologists to embrace statistical, Boltzmannian dynamics. Mendel’s generative role in these events looks prefigured in Depew and Weber’s treatment. They note his studies in physics at the University of Vienna in the early 1850s with Christian Doppler and other distinguished teachers, who thought well enough of Mendel’s performance to ask him to serve as a demonstrator at the Physical Institute. The legacies from this time, we learn, include: a sophisticated knowledge of experimental method; an openness to idealization; a mathematical competence that took in combinatorial analysis; and a familiarity with the atomistic conception of nature. (‘The idea that there are atoms of inheritance owes something’, write Depew and Weber, ‘to the Democritean imagination of modern physics.’) When Mendel took up questions about plant hybridization and evolution, as introduced by his Vienna botany teacher Franz Unger, he thus brought with him a physicist’s frame of mind.1

This image of Mendel as the physicist in the garden is far from idiosyncratic. Once the preserve of specialists, it now turns up in the pages of quality literary reviews.2In the meantime, however, the specialist literature has moved on. While it remains indubitable that Mendel took something away from his time in the physics classroom, the temptation to attribute to it everything conceptually or methodologically classy about his famous 1866 paper on his plant hybridization experiments should be resisted. As Sander Gliboff has now revealed, botany as Mendel learned it from Unger was already a physics-like science, born of two specifically Austro-German botanical traditions about as far removed from botany-as-plant-spotting as could be. One was the biogeography pioneered by Alexander von Humboldt, devoted to counting species and, with the aid of an array of instruments, measuring temperature, magnetic intensity, atmospheric composition and other physical variables in order to find out how the conditions in a place determine the range of species found there. The other, related tradition was comparative morphology, which, since Goethe (to whom Humboldt dedicated a key early work on plant geography), aimed to account for the diversity in living forms by relating them systematically to more basic underlying forms. Both traditions sought the laws of life, ideally quantitative laws such as, from morphology, Schimper’s law (1830), which states that the leaves spiralling up a stem are spaced at points characterized by ratios of alternate Fibonacci numbers—at 1/2 of a turn, or 1/3, or 2/5, or 3/8 and so on.3 Unger worked at the intersection, searching for the mathematical laws governing the changing distribution of plant forms as climate succeeded climate over geological time. ‘A physics of the plant organism’ was his phrase for what he wanted. Questions about the fixity of species, and the role of hybridization experiments in answering them, took their place in a comprehensive programme combining large-scale collection of quantitative data and big-picture theory. The values of precision here included the practical, since getting this science right had consequences for the wise management of the Habsburg Empire’s natural wealth. After the 1848 revolutions, reform-minded ministers channelled research funding accordingly.4

So, far from inventing it from scratch, Mendel inherited a botany that harnessed the idealizing, mathematical and reductive ways of physics to distinctively biological ends. Recalling these ends helps us understand the most abstract and apparently physics-like moment in his 1866 paper—a moment which, moreover, never features in conventional discussions of ‘Mendel’s laws’.5 Having led the reader through the now-familiar intellectual territory, Mendel concludes that plants in the second hybrid generation are distributed in the ratio 1A: 2Aa: 1a, where A is the dominant character and a the recessive character. He then generalizes that conclusion, asserting that, from the second hybrid generation onward, the plant forms are distributed in the ratio (2n−1)A: 2Aa: (2n−1)a, where n equals the number of generations.6 To later sensibilities, this all looks clever but surplus to requirements. Why expend so much effort on what happens after the second hybrid generation? Our textbooks lose the trail of descendants around there. After all, and as Mendel himself showed, one need only look to the parental, first hybrid and second hybrid generations to raise the questions for which the key Mendelian concepts—dominance, recessiveness, equal contributions from male and female, segregation, independent assortment—provide satisfying answers. Sensitive to the perplexity of modern readers of the paper, John Corcos and Floyd Monaghan, in their book-length guide, remark of this passage: ‘If this seems a strange way for a plant breederto reason, remember, Mendel was a very unusual breeder—he had been trained in physics and mathematics.’7 Gliboff interprets it differently. He suggests that what Mendel was after was not abstraction for its own sake, nor the ‘principles of heredity’ (William Bateson’s influential spin), but, in keeping with Unger’s project, a law that would predict, precisely, the long run of changing distributions in plant forms. With this ambition in view, Mendel’s generalization of the 1:2:1 ratio emerges as not just intelligible but, as Gliboff puts it, ‘the centrepiece of the paper’. The abstract rule showed that, down the generations, the proportion of hybrid plants will diminish by regular increments without ever fully disappearing. Yes, as the older experimental breeders had found, there will be reversion to the parental forms. But at the level of the trait-making factors, the hybrid form will remain. To that extent, hybridization leaves lineages permanently modified.8

It is a boon to have a previously dark corner of Mendel’s paper lit up so fully. And as Gliboff shows, the recovery of its specific botanical context does even more for the present-day student of Mendel, for instance by revealing his meteorological measurements to have been part of the same ‘Austro-Ungerian’ programme that inspired the hybridization research.9 Here I wish to extend Gliboff’s achievement in a new direction. If physics matters less than we once thought to the making of Mendel, it matters more than we tend to recognize to the making of Mendelism. That was, in its earliest phase, a British business. After the revival of interest in Mendel’s paper from 1900, his first and greatest champion was William Bateson, then based at Cambridge. And the Mendelian Bateson’s first and greatest critic was a former friend, the Oxford zoologist Walter Frank Raphael Weldon. Both men were steeped in a homegrown tradition in quantitative biology that started with Francis Galton, and their divergent responses to Mendel’s work can be traced to their attachments to distinctive parts of that tradition: in Bateson’s case, to the use of experiment (it was Galton who first encouraged him to study inheritance in experimental lineages of domesticated plants and animals) and the discontinuous nature of evolution; in Weldon’s case, to exact statistical description and the law of ancestral heredity. So much is well known. Less often noted—indeed, in Weldon’s case, never so—is the way that both drew upon the physics of their place and period in developing and publicizing their views. I want to show in what follows how attention to the role of physics in their polemics can help us better understand what was at stake in the ‘biometrician-Mendelian debate’ they fomented. In prospect is a richer sense not only of what each wanted for the scientific study of heredity circa 1900 but of the options available within the Galtonian tradition. At the end I shall draw a moral or two for students of the physics-biology nexus more generally.

2. Francis Galton and the molecular physics of germ structure

At the start, however, we need to look back briefly at Galton himself, and the consequences of his own most prolonged encounter with the putative lessons of physics for heredity. Throughout the 1870s, Galton shrugged off criticism of his physiological theory of inheritance from none other than James Clerk Maxwell. The exchange between Galton and Maxwell receives no mention in the most recent scholarly biography of Galton.10 But it merits scrutiny, as a reminder of the very long association between the sciences of heredity and concerns about ‘biological determinism’, and also of the remarkable fertility of Galton’s ideas among his contemporaries. Galton’s theory—which little occupied him thereafter—antagonized Maxwell into some of his most sustained reflections on the ultimate nature of reality.11 Thirty years later, the theory became, in Weldon’s hands, the basis for a biology concerned, as Mendelism so manifestly was not, with the complex environmental conditioning of inherited traits.

Although historians have noticed individual moments in the Galton-Maxwell dialogue, none before has given us the whole of it, or identified what seems to be Galton’s most direct rebuttal of Maxwell’s position. The opener was the appearance in June 1872 of Galton’s essay ‘On blood-relationship’, setting out views that would reach a wider audience in his epochal book Natural inheritance (1889). According to Galton, the ‘impregnated ovum’ is ‘structureless’, though containing materials that differentiate into two kinds: ‘patent’ or ‘personal elements’, which develop into the characters constituting the adult; and ‘latent elements’, which remain undeveloped but can be transmitted to offspring. The materials that become the patent elements are ‘segregated’ (his word) from the rest by a competitive process that selects out a representative subset. Differences between individuals in a lineage can thus be understood as arising from different selections made from the same set of inherited materials. Hence the ‘intellectual and moral gifts’ of individuals, as Galton wrote in conclusion, were not the less inherited for being variable.12

Published in the Proceedings of the Royal Society, the essay soon caught Maxwell’s interest. His fullest response came in a famous 1875 article in the Encyclopaedia Britannica under the title ‘Atom’. There were at least two earlier engagements, however. The first was in November 1872, at a meeting of the Cambridge Philosophical Society, where Maxwell offered a comment on the theory of pangenesis or ‘gemmules’—Darwin’s word for the particles that he postulated to account for heredity. Galton did not use ‘gemmules’ in his essay; but his ‘elements’ bore an obvious family resemblance, and in any case he had just concluded a protracted series of experimental tests of Darwin’s proposal. According to the Society’s Proceedings, Maxwell ‘spoke of the difficulty of conceiving chemical molecules in sufficient quantity being packed in these small gemmules’.13 A few months later, on 11 February 1873, he considered Galton’s position from a different point of view in an essay delivered to the Eranus Club with the title ‘Does the progress of physical science tend to give any advantage to the opinion of necessity (or determinism) over that of the contingency of events and the freedom of the will?’ Tongue firmly in cheek, Maxwell wondered whether, in the light of Galton’s latest contribution, one’s preference in the freedom-versus-determinism debate is itself hereditarily determined. On the possibility that it was not so determined, and therefore that ideas from physics might make a difference to what one thought, Maxwell went on to assess them, concluding that the world would seem less rigidly determined if we attended more systematically to how tiny influences can have large effects. He gave as an example ‘the little gemmule which makes us philosophers or idiots’.14

In the ‘Atom’ article he developed more fully the argument put to the Cambridge Philosophical Society. After introducing the major features of the modern theory of molecular gases and the experimental means used to estimate the size of molecules, he paused to make vivid just how small molecules are. Consider, he wrote, that the smallest cube we can see under a microscope—a cube whose edges are each 1/400,000th of a millimetre long—contains up to 100 million oxygen-sized molecules. He next put that eye-poppingly big number to work. What, he went on, if that cube were a cube of ‘organised substances’? One might think that a smallest visible cube of living stuff would contain around two million molecules, since each molecule of a living thing is made up of around 50 roughly oxygen-sized atoms or molecules. But two million is too many, Maxwell argued, since organisms are half water. The best estimate, then, is a million molecules—which number can also serve as an estimate, he reckoned, of the molecules in an ‘exceedingly simple organism’. Swiftly equating simple organisms with the germs from which complex organisms arise, Maxwell now argued that it is, however, ‘impossible … to conceive so small a number sufficient to form a being furnished with a whole system of specialised organs’. To show how, in his words, ‘molecular science sets us face to face with physiological theories’, he next summarized the theory of the germ as all-determining, not just of an individual’s organs and habits but, thanks to ‘a stock of latent gemmules’, those of even distant, and very different, descendants. Then came the knife:

Some of the exponents of this theory of heredity have attempted to elude the difficulty of placing a whole world of wonders within a body so small and so devoid of visible structure as a germ, by using the phrase structureless germ [here he cited Galton’s 1872 essay]. Now, one material system can differ from another only in the configuration and motion which it has at a given instant. To explain differences of function and development of a germ without assuming differences of structure is, therefore, to admit that the properties of a germ are not those of a purely material system.15

Ever since the physicist John Tyndall’s notorious Belfast Address of 1874, the need for Christian men of science of Maxwell’s calibre to defend science from materialism and associated doctrines had become more pressing. The ‘Atom’ article rendered service several times over. As Maxwell presented it, the new vortex theory of the atom—to be discussed at greater length below in connection with Bateson—made the determinist billiard-ball atomism underpinning Tyndall’s picture of nature look hopelessly out of date. Tyndall’s evolutionism also suffered a blow, with Maxwell reporting that the uniformity revealed through use of the spectroscope showed that atoms and molecules had been created, not evolved. (‘The constitution of an atom … is such as to render it, so far as we can judge, independent of all the dangers arising from the struggle for existence.’) As for Galton, he had been shown ignorant of the strictures that precision physics laid down on talk of structureless-but-material germs. Germs developing into different kinds of organism have to differ structurally, but do not contain enough molecules to support such structural difference. Hence something non-material must be involved.16

Later that year Galton published what appears to be a response to Maxwell. It is not so much an answer as a postponement wrapped in a put-down. In ‘A theory of heredity’, which came out in December 1875, Galton presented organic-unit theories of heredity such as Darwin’s and his own as the true inheritors of the mantle of ‘molecular’ science. (To the best of my knowledge, the word ‘molecular’ in the Maxwellian sense appears nowhere else in Galton’s voluminous corpus.) Such theories, he explained, insisted that organization ‘wholly depends on the mutual affinities and repulsions of the separate germs’. To believe otherwise—in what he called ‘a general plastic force’—was to believe in something that, for Galton, resembled ‘other mystic conceptions current in the early stages of many branches of physical science, all of which yielded to molecular views, as knowledge increased.’17

Undeterred, Maxwell continued to include Galton in their debate. ‘Do you take any interest in Fixt Fate, Free Will & c.?’ Maxwell wondered in a letter to Galton early in 1879. The letter went on to describe recent work that had done a great deal to show how material systems could be less than wholly determinate without, in Maxwell’s words, ‘the insinuation that there is something loose about the laws of nature, not of sensible magnitude but enough to bring her round in time.’ Although Galton’s polite reply suggests that he took no such interest, the letter may well have set him thinking.18 A few years later, as he recalled in his autobiography, ‘I was so harassed with the old question of Determinism, which would leave every human action under the control of Heredity and Environment, that I made a series of observations on the actions of my own mind in relation to Free Will.’ His finding, reported in Mind in 1884, was that the more closely he inquired into his own acts of apparent free will, the more clearly he saw those acts as bound within ordinary causal chains. Galton—no Christian—later summed up his results as supporting ‘the views of those who hold that man is little more than a conscious machine, the slave of heredity and environment, the larger part, perhaps, of whose actions are therefore predictable.’19

3. W. F. R. Weldon and the methodological virtue of precision physics

What startles in the above is not Galton’s determinism but his letting environment in on the act. We remember him as the arch hereditarian. But that was not how he saw himself.20 Nor was it how Weldon, Linacre Professor of Zoology at Oxford and the most accomplished defender of Galton’s theory of heredity, understood him. I shall return below to Galton’s environmentalism. I want first to consider a couple of episodes from the precision physics of the early 1890s and why, in the mid-1900s, Weldon dwelt upon them in his attempts to win converts to the Galtonian-environmentalist cause.

The first episode comes up in the opening chapter of Weldon’s Theory of inheritance, which he worked on from the summer of 1904 to the summer of 1905, but left unpublished, and incomplete, at his death in 1906. Here Weldon introduced statistical description and its power to record experience with exactness. By way of illustrating the risks inherent in non-statistical description—in, that is, the use of an average or otherwise representative value in place of the full range of observations made or measurements obtained—he turned to the problem of latitude in astronomy. Look up the latitude of the telescope of the Radcliffe Observatory at Oxford, and, he reported, you will find a range of values, all of them extremely close to the single figure generally given as the latitude. And yet, he continued, it was only by taking such apparently trifling variability seriously that astronomers discovered that latitude is not, in fact, absolutely constant, but varies, changing with periodic changes in the position within the Earth of the axis of spin (and therefore in the position of the Equator). A sensational revision in the understanding of the Earth as a physical system thus followed from a refusal to sweep variability under the carpet of make-believe constancy.21

Weldon returned to the latitude example, and the lesson he drew from it, in a lecture on inheritance in August 1905 at Oxford, presented as his contribution to an omnibus series there on ‘the method of science’. At the end of the latitude discovery story he now added that whenever men substitute a constant value for a variable record carefully assembled, it is always legitimate to ask whether they are justified in doing so. Was the variability experienced due to their own limitations (in which case the substitution is warranted)? Or was it due to genuine differences in the thing under study? ‘It is just this region of uncertainty, fringing our experience with a haze of doubt’, commented Weldon, ‘which fascinates the real experimenter. He is always striving to reduce it within narrower limits, and to enlarge the region of experimental certainty as much as he can.’ To drive the point home, Weldon told a further discovery story of the same vintage, this one about the elements. Attempting to determine the weight of nitrogen, the physicist Lord Rayleigh had found that when he weighed a volume of it derived by extracting all known constituents except nitrogen from atmospheric air, he got a set of measurements that were fractionally higher—on the order of hundredths of a gram—than the ones he got when he weighed nitrogen derived from nitrogen-containing compounds. He concluded that atmospheric air must contain hitherto unknown constituents. These were soon identified as the inert gases, chief among them the element known ever since as argon. Weldon went counterfactual to emphasize how much science owed to a skilful experimenter’s unwillingness not to let himself off the hook when confronted with contradictory data. ‘If Lord Rayleigh had replaced all the results in his table by a single compromise between them, and had been content to stop there’, Weldon concluded, ‘we should not know the existence of argon to-day.’22

And then there was Mendel. In the Oxford lecture Weldon did not make the unflattering comparison explicitly, instead turning from the argon story directly to the Galtonian statistical apparatus that, as he saw it, promised to help biologistscontend with the whole of their results on heredity, not just an idealized version of them. To grasp the anti-Mendelian point of Weldon’s tour of recent precision physics we need to go back to the Theory of inheritance manuscript. All becomes clear in the second chapter, devoted to comparing and contrasting Galton’s and Mendel’s theories of heredity. The discussion of the latter is not disrespectful. But there is no mistaking where Weldon’s sympathies lie. In his view, the root problem with Mendel’s work on hybrid peas is his descriptive categories, which impose a false constancy on variable reality. Pea seeds are not uniformly ‘green’ or uniformly ‘yellow’, uniformly ‘round’ or uniformly ‘wrinkled’. As shown in photographic plates accompanying Weldon’s first article on Mendel in 1902 (thus rousing Bateson to publish his notorious Mendel’s principles of heredity: A defence later that year), seeds can be grouped into green-to-yellow continua and shown to exhibit different degrees of wrinkledness. Allowing for such differences, one can even notice, for example, how much more closely the wrinkles of a given variety resemble those of a distant ancestor than an immediate one. Such patterns were lost to the Mendelian, enchanted by a theory erected on the basis of variability-masking categories. Of course, Weldon explained, in working with parental pea varieties that had been heavily inbred, Mendel would have seen a lot more uniformity of character in the hybrid offspring than he otherwise would have. But his decision to ignore even such variability as persisted by describing all the seeds in the first hybrid generations as ‘yellow’ and ‘round’ opened the door to his distinctive—and flawed—concept of dominance. For Mendel, yellow is dominant to green and round dominant to wrinkled, full stop. Here was dominance as, in Weldon’s words, ‘something permanently associated with one character of each alternative pair’. And from dominance so conceived it was a short step to segregation understood as a purity-enforcing mechanism; for if, in the second hybrid generations, there appeared (as there did) green and wrinkled seeds, then those must have arisen from gametescontaining only green-making and wrinkled-making determinants—in which case, concluded Mendel, the hybrids must manufacture gametes that are pure for one or the other of a pair of characters.23

The contrast with Galton’s concept of dominance was stark. Where Mendel had treated dominance as sublimely independent of context—yellow is visible if a yellow-making determinant is present, no matter what company it keeps—Galton had treated it as radically, and inclusively, context-dependent. Wrote Weldon:

Galton, …, observing the results of mating which occurred very nearly at random, found that the same character of an alternative pair might be at one time dominant, at another recessive, so that he was led to regard dominance as depending not upon the character borne by a germinal determinant, but upon the condition of the determinant itself, and upon its relation to the other germinal determinants, at some moment during fertilisation or subsequent development.24

Accordingly, Weldon explained, if you cast a Galtonian eye over Mendel’s results, you will find yourself asking not just about overlooked variation but about ‘the peculiar conditions of the experiment’ which served, for a while, to keep yellow, round and the rest so overwhelmingly dominant.25 Weldon’s source for this environmentalist reading of Galton was none other than the 1872 paper that had so distressed Maxwell. For Weldon, this paper was the foundation of the Galtonian edifice. Here was the physiological theory whose fruitful redirections of attention—toward the whole of a population and its history for patterns of dominance, toward the throwing of dice as a model for relations between generations—had culminated in the famous Galton-Pearson law of ancestral heredity. And at the theory’s core was the view that hereditary determinants dominate when they do thanks to causal interactions that, like those involved in Darwinian competition, are multiple, complex and ever-shifting.26

In 1872 Galton had reached for a political analogy to clarify the nature of the cell-level competition he envisaged.27 Given the knowledge of cell structure at the time he could hardly have done more, wrote Weldon. But now things were different, and ‘we shall see in a future chapter how far the suggestion made can be brought into harmony with the facts of germinal structure which have been discovered during the last thirty years’.28 Alas Weldon never got to write that chapter, though surviving manuscripts indicate something of his ideas about the chromosomal underpinnings of Galtonian heredity.29 The rest of the book mostly provides an overview of recent experimental studies of growth and regeneration, starting with single-celled organisms, then on to complex invertebrates, and ultimately up to vertebrates. What these studies showed, on Weldon’s interpretation, is that biological tissues reveal different potentialities depending on their surroundings. Change the surroundings and you may well change what develops. Above the level of the hereditary determinants, then, dominance—understood as the state of being patent rather than latent—is conditional. Biologists should want a theory of hereditary dominance that respects this finding. And that theory was Galton’s, not Mendel’s.30

In one important sense, Mendel’s work was too precise for Weldon. What we now describe as the ‘Mendel-Fisher controversy’, over whether Mendel’s data fitted his theory improbably closely, began as the Mendel-Weldon controversy. The same 1902 paper that included those plates of peas also featured the first statistical inquiry into the chances of Mendel having observed as high a proportion of theory-confirming plants as he reported.31 But this analysis of, as Weldon put it, ‘the wonderfully consistent way in which Mendel’s results agree with his theory’ was by no means the main point of Weldon’s article. He did not even mention it in his discussion of Mendel in the Theory of inheritance.32 The main point was the one that the plates illustrated: Mendel’s imprecise categories do not well describe variable experience. On its own, that could seem—and did seem to Bateson—a nit-picking complaint.33By the time of Weldon’s 1905 Oxford lecture, however, he had hit upon the tales from physics that made vivid the stakes involved. (In 1904 Rayleigh had won the Nobel Prize in physics for his part in the discovery of argon; that same year Weldon’s Oxford colleague and statistical collaborator H. H. Turner, Savilian Professor of Astronomy, published a popular book, Astronomical discoveries, recounting the latitude story.)34 In dramatizing how inexact Mendelism was and how that inexactness would handicap the science of heredity, Weldon thus fulfilled the description of him in Bateson’s Defence as ‘a vehement preacher of precision’.35

4. William Bateson and the biology of the vortex

Mendel wrote before the discovery of chromosomes. But his notion of segregation—for Mendel, the idea that hybrids form equal numbers of A-determining gametesand a-determining ones—turned out, of course, to lend itself remarkably well to chromosomal glossing.36 Bateson, as the Mendelian-in-chief for the first decade of the twentieth century, always held out hope for a more fundamental explanation. He wanted the Mendelian pattern to take its place in a comprehensive theory of organic patterning, with chromosomes treated as cellular elements that, like other such elements, come to be coordinated into pattern-generating movement. To concentrate solely on the likes of chromosomes was to close off any possibility of understanding the coordinating forces. ‘We can study the processes of fertilisationand development in the finest detail which the microscope manifests to us’, wrote Bateson in 1900, in his first Mendelian publication (reproduced in the 1902 Defence), ‘and we may fairly say that we have now a thorough grasp of the visible phenomena; but of the nature of the physical basis of heredity we have no conception at all.’ What was needed, he continued, was the ‘suggestion, working hypothesis, or mental picture’ that would enable biologists ‘to penetrate beyond what we see’. But it was nowhere to be found. He drew a parallel: ‘We are in the state in which the students of physical science were in the period when it was open to anyone to believe that heat was a material substance or not, as he chose.’37

To succeed as physicists had, Bateson implied, biologists would need to break free from substance explanations. Heat had ceased to be mysterious once it was understood not as the distinctive property of a material (caloric) but as a generic property of moving particles of any material. Heat was ‘a mode of motion’, as John Tyndall had put it in the title of an 1863 bestseller. So was the pressure of gases. Matter itself looked for a time as if it would yield to mechanical analysis. Under the theory developed and popularized in the late 1860s and 1870s by the Scottish physicists P. G. Tait, William Thomson and Maxwell (from initial studies by the brilliant Hermann von Helmholtz), atoms were pictured not as material particles but vortices in a universally distributed ether. Their whirling accounted for their hardness and permanence, on the model of the spinning smoke rings that, as Tait demonstrated experimentally (with Thomson in the audience), bounced off each other when they collided. The ‘encounter between two smoke rings in air’, affirmed Maxwell in his 1875 encyclopaedia article on the atom, ‘gives a very lively illustration of the elasticity of vortex rings.’ Smoke rings became standard fare in the lecture theatre as well as on the page. At the height of the vortex theory’s popularity in Britain, between roughly 1875 and 1885, it promised to unify everything from gravity to chemistry. But the difficulties mounted and enthusiasm waned. The rest of the world was aloof from start to finish. Yet even into the early twentieth century, the theory’s admirers held it up as a shining example of the mechanical view of nature that British science had served so well.38

Educated in the natural sciences at Cambridge between 1879 and 1883, during the heyday of vortex physics, Bateson to the end of his life carried a torch for vortex biology. William Coleman pieced together the documentary evidence in a pioneering article forty years ago.39 A living thing, wrote Bateson in a letter of 1924, ‘is not matter. It is a system-vortex … through which matter is passing.’40 In an address that same year, he defined biology as the study of such vortices.41 When he tried to give a sense of how differently the vortex biologist looked at a cell or a body part or an organism, he reached for simple physical phenomena showing how vibrations can organize matter into form. Chladni patterns were a particular favourite: pour sand on a plate, set it vibrating with sound or a violin bow, and the sand will gradually clear from the high-motion places and clump in the low-motion ones, forming different patterns according to the frequency of the vibration. Wind-rippled beaches and octave-apart piano wires were other staple analogues. Discussion of them shows up in Bateson’s private correspondence and published writings from the early 1890s.42 But his interest in this way of thinking went further back. He credited first stirrings—though Coleman omitted mention of it—to Michael Foster, boss of laboratory physiology at Cambridge when Bateson was an undergraduate there. He once wrote that the ‘best answer in few words’ that he knew to the question, ‘what is a living thing?’,

is one which my old teacher, Michael Foster, used to give in his lectures introductory to biology. ‘A living thing is a vortex of chemical and molecular change.’ This description gives much, if not all, that is the essence of life. The living thing is unlike ordinary matter in the fact that, through it, matter is always passing. Matter is essential to it; but, provided that the flow in and out is unimpeded, the life-process can go on so far as we know indefinitely. Yet the living ‘vortex’ differs from all others in the fact that it can divide and throw off other ‘vortices’, through which again matter swirls.

How might such living vortices replicate? Like twisting smoke rings, of course.43

When we consider what Bateson the vortex biologist found so exciting about Mendel’s paper, three attractions stand out. First, Mendel’s concern with unit characters—greenness or yellowness with no shadings in between—was a good match for Bateson’s conviction that the most illuminating patterns in biology were, like the resonant frequencies of Chladni-pattern plates, discontinuous. Although Bateson had biological reasons for favouring discontinuity, these cannot be easily separated from the physical considerations described above.44 Second, Mendelian segregation seemed to Bateson a marvellous example of what he called a ‘meristic force’—a part-distributing force, acting in the same way regardless of the material composition of what is distributed. It makes no difference that the chemicals responsible for colour alternatives in the seeds of hybrid pea plants are different from those responsible for height alternatives in those plants; segregation, a mechanical process, operates the same.45 Third, although Mendel tracked his characters through lineages of organisms, there was nothing in principle, Bateson thought, to forbid extension of the analysis to lineages within organisms—lineages of cells, say, or more complex parts such as leaves or vertebrae. He would go on to discuss evidence of exactly this kind of segregation. For him, a theory of biological pattern worth having had to deal with pattern formation at all scales. On this score too, Mendel’s approach looked promising.46

The quantitative exactness of that approach was, to be sure, another attraction for Bateson. No hybridization experiment before, Mendel had reckoned, and in the words of his 1866 paper, had ‘been carried out to such an extent and in such a way as to make it possible to determine the number of different forms under which the offspring of hybrids appear, or to arrange these forms with certainty according to their separate generations, or to definitely ascertain their statistical relations’. Bateson underscored the far-reaching nature of these breaks with precedent. ‘It is to the clear conception of these three primary necessities that the whole success of Mendel’s work is due’, he wrote in an appreciative footnote appended to the translation of the paper in the Defence. ‘So far as I know this conception was absolutely new in his day.’47

Was Bateson’s valuing of Mendel’s fastidiousness in these matters also a legacy of vortex physics? Not likely. For one thing, measurement never became a key part of the vortex-atom programme, which started and remained a largely mathematical endeavour.48 For another, Bateson, like Mendel, was heir to native traditions in quantitative biology. I have already mentioned Foster, on whose courses in biology Cambridge beginners learned for themselves the arts of exact measurement with delicate apparatus.49 Later, and more famously, Bateson aligned himself with Galton, whose Natural inheritance showed how to conduct statistical investigations of heredity and even how to think in physical terms about discontinuous evolution (Galton’s image was a polygon, stable when resting on its faces but otherwise unstable).50 Bateson looked up to physicists as masters not of precision measurement but of dynamical analysis. In a 1917 lecture in Cambridge, he recalled once again what Foster used to say about living things, adding: ‘If we could in any real way identify or analyse the causation of growth, biology would become a branch of physics. Till then we are merely collecting diagrams which some day the physicist will interpret.’51

I have noted Foster’s absence from Coleman’s account of Bateson’s vortex biology and its sources. There is another, more consequential problem with that account. Coleman lumped unhelpfully when he wrote that Maxwell’s ‘attempted destruction of the notion of a material or particulate basis of inheritance represents … the beginning of the opposition party which Bateson so splendidly illustrates’.52 As we have seen, Maxwell’s ‘Atom’ article included an attack on Galton’s materialist theory of heredity and a précis of the vortex theory of matter; but the former in no way drew upon the latter, which is what bound Bateson to Maxwell; and Maxwell’s notion of a more-than-material basis for heredity, and the Christian yearnings it expressed, have as little to do with Bateson’s understanding of the Mendelian mechanism as with the chromosomal understanding that he opposed. More needs to be done to unpick the Coleman package.53

5. A Fisherian postscript – and a moral or two

I began by recalling Depew and Weber’s Mendel, whose hereditary factors ultimately made possible the reformulation of natural selection theory along Boltzmannian lines. That was the work above all of Ronald Fisher. If I do not dwell here on Fisher’s position in this story of physics in the Galtonian tradition, it is because Jonathan Hodge has made that position so conspicuous. Unlike Galton, Weldon or Bateson, Fisher trained in a serious way in physics, even spending a postgraduate year in Cambridge under James Jeans. Statistical thermodynamics became the model science for Fisher, who, in The genetical theory of natural selection (1930), drew explicit comparisons and contrasts between the second law of thermodynamics and his own fundamental theorem of natural selection. Both concerned transitions between less probable and more probable states in vast ensembles. But where the erosion of energy gradients in molecules brought about entropic decline, the erosion of fitness gradients in genes brought about evolutionary progress. Fisher’s universe was thus, as Hodge says, a ‘two-tendency’ universe. Yet for all that it was a place congenial to eugenical hopes, and all that Fisher drew on Galton’s work in the avowedly eugenical chapters in the 1930 book, the descendant shared none of the ancestor’s determinism. A devout Christian, Fisher sided with Maxwell: physics taught, and biology affirmed, indeterminism.54

How much diversity the Galtonian tradition housed! Its proponents busied themselves with very different kinds of biological research whilst connecting that research in remarkably different ways with very different kinds of physics. One way to differentiate what I refer to in my title as ‘the Galtonian sciences of heredity’ is to ask about those connections with physics. Galton heaped scorn on Maxwell’s views on matter and determinism in developing an influential vision of how individuals come to exhibit, and to pass on, the hereditary potentialities they do. Weldon, who subscribed to that vision, took it to endorse a radical environmentalism: hereditary ‘determinants’ had to be understood as having their effects in conjunction with the effects of lots of other causes, some of them within the body, some of them not. Chastising the Mendelian Bateson for idealizing the messy data on inheritance, Weldon turned for support to physics—not, however, to uniformity-prizing Maxwellian matter theory, but to variation-prizing geodetic astronomy and atomic-weight studies. Bateson does not seem to have taken over any of Galton’s environmentalism. What Bateson took from Galton—and what made Mendel’s work so appealing—was the emphasis on discontinuous change. Mendelism as Bateson articulated it was a hereditarian science, tracking atomistic traits that marched through the generations. In making sense of biological discontinuity, however, Bateson drew from the conceptual world of the vortex atom which Maxwell had done so much to promote. (Bateson’s friend and colleague Reginald Punnett in his Mendelism (1909) celebrated the theme of ‘Discontinuity—discontinuity of the atom, and the discontinuity in living forms’, evoking a future where atomism would advance the science of the living as it had earlier advanced the science of the nonliving.)55 And then there was Fisher, who found in statistical thermodynamics a resource capable of binding together Galton’s eugenics, Weldon and Pearson’s Darwinian biometry, Bateson’s Mendelism, and his own Christian belief in free will.56

Physics, in its many varieties, helped the Galtonians rethink their tradition. It can also help those of us concerned with understanding that tradition historically to rethink our own traditions, by helping us recover all the flexibility familiar to practitioners. A history of the biometrician-Mendelian controversies which takes seriously that flexibility—and takes seriously as well its gradual disappearance, as Mendelism came to absorb or otherwise neutralize the criticisms of the likes of Weldon, and Weldon’s vision of biometry came to be displaced by something far less resourceful—will look very different from the ones we now have. That is a lesson for historians of biology.57 But it points up a larger moral, of interest to all concerned with what happens when physics meets biology. A venerable contrast, still plenty current (not least in the introduction to this special issue), holds that where physicists idealize, mathematize and reduce, biologists revel in diversity and particularity. Yet as we have seen, in the same century in which ‘physics’ and ‘biology’ first acquired those names, they were already messily heterogeneous and overlapping. There were idealizing, mathematizing and reducing biologies, and there were branches of physics hung up on the tiniest, generalization-frustrating particulars. What matters, then, for any given physics-biology encounter, is less that some kind of physics met some kind of biology, than that this kind of physics met that kind of biology.

I conclude, then, with a plea for pluralism. It is none the worse for being, in our times in history and philosophy of science, quite conventional. Nor is it meant to suggest that there are no worthwhile questions to be posed at higher levels of analysis. To take only the most obvious: granted all this motley, how and why did the situation come to be remembered so differently? One might think the blame rests with disciplinarily chauvinistic physicists and biologists; and no doubt these have played their part.58 But, as Simon Schaffer has recently argued, it is often those seeking to transcend disciplinary barriers, through the creation of ‘interdisciplines’, who do a lot of the representational heavy-lifting. In making room for a new interdisciplinary science, its promoters reliably canonize disorderly constituent sciences as orderly disciplines which now need to be in conversation.59 This special issue can be understood as a case in point. Its origins lie in the efforts of our scientific prime movers, Tom McLeish and Wilson Poon, to bring the physics of complex fluids—‘soft condensed matter physics’—to bear on biological problems. The phrase ‘physics meets biology’ earlier appeared in an article they wrote advertising the interdisciplinary potential of their science. That began: ‘The links between physics and biology have become increasingly close in recent years’.60 It would, as we have seen, be nearer the truth to say that the links between physics and biology are perpetually announced as increasingly close. And as long as scientific imaginations continue to be fired by the prospect of forging those links, they will continue to impart stability to the otherwise unstable identities of physics and biology.

Acknowledgements

Many thanks to Tom McLeish, Wilson Poon and Alexander Bird for inviting this paper; to Darrell Rowbottom for extracting it; to two anonymous referees for vetting it; and to Berris Charnley, Donald Forsdyke, Jon Hodge, Evelyn Fox Keller, Annie Jamieson, Eileen Magnello, Staffan Müller-Wille, Chris Renwick, Simon Schaffer and everyone at the ‘Physics meets biology’ meeting, Edinburgh, November 2008 for much helpful discussion. Funding for my work on Weldon’s Theory of inheritance was provided by the British Academy.

Footnotes
1 Depew & Weber (1995), esp. pp. 217–223, quotation on p. 221. On the Depew-Weber thesis, see Radick (1998).
2 For a distinguished example of the debt-to-physics motif in the historical literature on Mendel, see Olby (1985), pp. 98–99. Among more popular writings, see, e.g., Berry (2001), which even suggests that the quantitative style imparted by Mendel’s physics training accounts for the long indifference to his work, as it baffled contemporary biologists whose science ‘was at the time almost wholly descriptive’ (p. 29).
3 Gliboff (1999), pp. 218–223. On Humboldtian science, see Dettelbach (1996). On the Goethean morphological tradition, see Richards (2008), appendix 1. On Schimper’s law, see Livio (2002), pp. 109–110 and, for historical background, Montgomery (1970), esp. pp. 306–311. The Fibonacci sequence is 1, 1, 2, 3, 5, 8 …, each term being the sum of the two preceding.
4 Gliboff (1999), pp. 218–225, quotation on p. 218. ‘Plant geography can be considered part of terrestrial physics’ was how Humboldt himself had put the point in 1820; quoted in Dettelbach (1996), p. 297. For more on Unger in Vienna, see Gliboff (1998).
5 From part 7, ‘The subsequent generations [bred] from the hybrids’; Mendel (1866), pp. 57–59. The paper is available on the web at www.mendelweb.org. So far as I know, there was no way empirically to separate out the Aand the Aa plants when Mendel wrote—in which case what he describes had the status of a thought experiment. It is worth noting too that Weldon, writing before understanding of Mendel’s paper had ‘hardened’, included the formula in his summary; Weldon (1902a), p. 231.
6 He reasons as follows. Suppose that we extract from the second hybrid generation the smallest set of plants that retains the 1:2:1 ratio. If we now let these four plants—one of the A form, two of the hybrid Aa form (outwardly identical to the A plants) and one of the a form—count as the first generation, then, with n equal to 1, the formula gives the correct numbers. Now suppose that each of these four plants self-fertilizes, and that each subsequently produces four seeds. The new generation of sixteen plants arising from these seeds will comprise: (1) from the Aparent, four A plants; (2) from one of the Aa parents, one A plant plus two Aa plants plus one a plant (as per the classic analysis); (3) from the other Aa parent, the same distribution as in (2); (4) from the a parent, four a plants. If, finally, we now group together all the A plants and all the a plants, we will find that we have 6 A plants, 4 Aa plants, and 6 a plants, giving a ratio of 3: 2: 3 or, with n equal to 2, 2n−1: 2: 2n−1. However large n is, the formula holds good.
7 Corcos & Monaghan (1993), p. 99.
8 Gliboff (1999), pp. 227–228, quotation on p. 227; Mendel (1866), pp. 58–59.
9 Gliboff (1999), pp. 218–219 and 223–225; for the phrase ‘Austro-Ungerian’, p. 223. On the links between meteorology, statistics and physics in Mendel’s formative years, see Olby (1985), pp. 98–99.
10 Gillham (2001)—though mention is made (p. 40) of Galton at Cambridge studying under the same illustrious mathematics coach, William Hopkins, who taught Maxwell.
11 On Galton’s putting the theory on ice, and his reasons for doing so, see Renwick (in press).
12 Galton (1872), quotations from pp. 395 & 402; cf. Galton (1889), pp. 18-19, ch. 11.
13 On Maxwell at the 1872 Cambridge meeting, see Peter Harman’s editorial note in Maxwell (1873), p. 815. On Darwin’s gemmules and the development of Galton’s ‘speculative physiology of heredity’, see Gayon (1998), pp. 106–115.
14 Maxwell (1873), quotation on p. 822; excerpted in Schaffer (1999), p. 32. I learned a great deal about Maxwell’s ‘Atom’ from Simon Schaffer’s primary-sources seminar on it in Cambridge, October 1999.
15 Maxwell (1875), pp. 460–461.
16 The quotation from Maxwell is in Maxwell (1875), p. 483. On the ‘Atom’ article in philosophical-ideological context, see Harman (1998), pp. 197–208; Kragh (2002), p. 86, affirms that the vortex atom ‘was not connected with the materialism and determinism that since the days of Democritus had been associated with atomism.’ So far as I know, Coleman (1970, p. 265) is the only historian of biology to treat Maxwell’s attack on Galton in the article.
17 Galton (1875), quotations on pp. 82-83. I first learned of this passage from Evelyn Fox Keller’s book on the nature/nurture debate, which she kindly let me read in manuscript; see Keller (2010). The search for ‘molecular’ in Galton’s writings was conducted at http://www.galton.org/ On the 1875 paper as published in December, see Galton (1876), p. 329.
18 Maxwell to Galton, 26 February 1879, in Harman (2002), pp. 756–758; for Galton’s reply, see p. 758, note 13. For discussion, see Harman (1998), pp. 205–208; also Porter (1986), pp. 205–208 (with a full transcription of Galton’s reply on p. 207, note 36) and Hacking (1990), pp. 155–156.
19 Galton (1908), quotations from pp. 295-296; cf. Galton (1884). Porter too (1986, p. 207, note 36) links Galton’s paper on free will to Maxwell’s 1879 letter.
20 In 1904 Galton defined eugenics as ‘the science which deals with all influences that improve the inborn qualities of a race; also with those that develop them to the utmost advantage.’ Galton (1904), p. 1, emphasis added. I’m grateful to Chris Renwick for helpful discussion of Galton on heredity/environment relations.
21 W. F. R. Weldon, Theory of inheritance (1904-5), ch. 1, esp. pp. 4–5, Box 264/2D, Papers of Karl Pearson, Special Collections, University College London (hereafter ‘Pearson Papers’). For the period when Weldon was working on the book, see Pearson (1906), pp. 44–46. On the discovery of the variation of latitude, see Carter & Carter (2002).
22 Weldon (1905), pp. 88–94, quotations on pp. 88, 92-93. I’m grateful to Annie Jamieson for alerting me to the importance of this lecture. On the discovery of the inert gases, see Brock (1992), pp. 331–337.
23 W. F. R. Weldon, Theory of inheritance (1904-5), ch. 2, quotation on p. 16, Box 264/2D, Pearson Papers; Weldon (1902a). See also Weldon (1902b) and, for discussion (including the quotation), Olby (1989a), pp. 315–316, though n.b. Olby counts the Galton-Mendel chapter—entitled ‘The theories of inheritance based on statistical observation’—as the third (the first, introductory chapter is present in alternate versions).
24 W. F. R. Weldon, Theory of inheritance (1904-5), ch. 2, quotation on p. 16, Box 264/2D, Pearson Papers; also in Olby, 1989a, Olby, 1989b.
25 W. F. R. Weldon, Theory of inheritance (1904-5), ch. 2, quotation on p. 21, Box 264/2D, Pearson Papers.
26 W. F. R. Weldon, Theory of inheritance (1904-5), ch. 2, esp. pp. 1-8, Box 264/2D, Pearson Papers. For the links between the Galtonian/Weismannian theory of the reduction division, see Weldon (1905), pp. 99–100. On the ancestral law, see Weldon (1905), p. 108 and, for historical analysis, Gayon (1998), pp. 136–144.
27 For Galton’s extended comparison between the competitive selection of embryonic elements and that of members of a ‘representative assembly’, see Galton (1872), p. 395.
28 W. F. R. Weldon, Theory of inheritance (1904-5), ch. 2, quotation on p. 3, Box 264/2D, Pearson Papers.
29 See Pearson (1908). On Pearson’s reconstruction of Weldon’s notes, he re-derived the classic Mendelian pattern by assumptions—notably about number of factors and chromosome breakup—other than Mendel’s, with the result that the ‘recessive’ offspring in the second hybrid generation harboured unexpressed factors for the dominant character.
30 W. F. R. Weldon, Theory of inheritance (1904-5), esp. ch. 2, pp. 21–23, Box 264/2D, Pearson Papers.
31 See Weldon (1902a), pp. 232–235. For comprehensive discussion of the controversy, see Franklin, Edwards, Fairbanks, Hartl, & Seidenfeld (2008); on Weldon’s role, see pp. 16-18.
32 Quotation from Weldon (1902a), p. 232.
33 Near the end of his Defence Bateson thundered: ‘The suggestion amounts to this: that because there are some exceptions to dominance in peas; and because by some stupendous coincidence, or still more amazing incompetence, a bungler might have thought he found dominance of one eye-colour whereas really there was none; therefore Professor Weldon is at liberty to suggest that there is a fair chance that Mendel and all who have followed him have either been the victims of this preposterous coincidence not once, but again and again; or else persisted in the same egregious and perfectly gratuitous blunder. Professor Weldon is gifted in the Calculus of Chance; will he compute the probabilities in favour of his hypothesis?’ Bateson (1902), pp. 192–193.
34 On Rayleigh’s Nobel Prize, see the entry on John William Strutt (Third Baron Rayleigh) in the Dictionary of Scientific Biography (old series), vol. 13, p. 103. For Turner’s version of the latitude variation story, see Turner (1904), ch. 6. In the statistical appendices following the introductory chapter of the Theory of inheritance, Weldon thanked Turner for help with one of the proofs; W. F. R. Weldon, Theory of inheritance (1904-5), appendix 1 to ch. 1, p. 1, Box 264/2D, Pearson Papers. When Weldon died, Turner wrote a letter to the Oxford Magazine describing his friend as a biological Tycho, dedicated to gathering ‘patient and laborious measurements before others had realized the necessity for them’; letter from H. H. Turner, Oxford Magazine, 31 Oct. 1906, clipping in Box 259/6, Pearson Papers.
35 Bateson (1902), p. xi. On an earlier controversy in which Weldon expressed much the same view about scientific method, see Pence (forthcoming).
36 On the history and prehistory of the chromosomal theory of the gene, see Schwartz (2008). Although Walter Sutton in 1902 is usually credited as the first to see the potential for bringing the chromosomal and Mendelian theories of inheritance together, the perception of overlap is there in the pages of Carl Correns’s 1900 ‘rediscovery’ paper. He noted that the Mendelian picture of a 1:1 ratio of dominant and recessive factors in the gametes of hybrid plants ‘strongly suggests that the separation occurs during a nuclear division, the reduction division of Weismann’—this at a time when the nucleus-based chromosomes were already widely suspected of playing a major role in heredity. Quotation from Correns (1900), p. 127, emphasis in original; discussed in Gliboff (1999), p. 230. But for a stern warning against reading too much into Correns’s statement, given the lack of consensus in 1900 about what would soon be called ‘meiosis’, see Baxter & Farley (1979).
37 Bateson (1900), p. 172; with minor changes in Bateson (1902), p. 3. Although no secondary source is authoritative on Bateson on chromosomes, much can be learned from Coleman (1970), the critical response in Cock (1983) and the extensive overview in Cock & Forsdyke (2008), ch. 13. For a statement of Bateson’s views as the chromosomal gene theory was hitting its stride, see Bateson (1913a), pp. 270–272.
38 For an overview of the vortex theory by a historically minded contemporary, see Merz (1965), pp. 56–66; see esp. p. 57 on Tyndall’s book and p. 63, note 1, on the public life of smoke rings (‘Smoke-rings, solid and liquid gyrostats, and a host of similar contrivances, have impressed on us the hidden resources of whirling motion.’). It is to Merz that I owe the phrase ‘mechanical view of nature’. For an overview by a culturally minded historian of physics, emphasizing the vortex theory’s status as a ‘Victorian theory of everything’, see Kragh (2002). The quotation from Maxwell is in Maxwell (1875), p. 472.
39 Coleman (1970). Although it remains an indispensable guide, the article is problematic for its misconstrual of Bateson’s attitude to organic form as in a strong sense ‘anti-materialist’. For a start on the needed critique, see Cock (1983) and the final paragraph of this section. For more extensive treatment, see Radick (forthcoming).
40 Bateson to F. B. Borradaille, 28 January 1924, cited in Coleman (1970), p. 270 and quoted in Schaffer (2000).
41 Bateson (1924), p. 408.
42 See esp. Bateson’s famous letter to his sister Anna of 14 September 1891 about what he regarded as ‘the best idea I ever had or am likely to have’; in B. Bateson (1984), pp. 42–43. The other letters collected on pp. 44-46 together with the editorial comments are also instructive, as is Bateson’s undated and unpublished manuscript ‘A “vibratory theory” of linear and radial segmentation as found in living bodies’, which exists in two slightly different copies in the Papers of William Bateson, Cambridge University Library, folder A.9.c. Bateson was sensitive to the limitations of thinking with Chladni patterns; see Bateson (1913b), p. 60 on how heaped sand does not go on to differentiate chemically as would divided living tissue. For a marvellously clear discussion of Chladni patterns as models of Batesonian morphogenesis, see Newman (2007), esp. pp. 88–90.
43 Bateson (1913b), pp. 39–40. Sometimes Bateson commented that Foster’s definition adapted a passage in Cuvier (see Bateson (1917), p. 209); sometimes Cuvier alone was mentioned as the source (see the 1924 letter to Borradaille, cited in note 40 above). The relevant quotation from Cuvier can be found in Cuvier & Latreille (1831), p. 7 (‘Life then is a vortex, more or less rapid, more or less complicated, the direction of which is invariable, and which always carries along molecules of similar kinds, but into which individual molecules are continually entering, and from which they are continually departing, so that the form of a living body is more essential to it than its matter.’—emphases in original). I’m grateful to Jon Hodge for discussion of Cuvier’s text.
44 In Materials for the study of variation (1894), Bateson argued that his concern with discontinuous variation was justified both because the origin of species appeared to be a discontinuous process (pp. 17–18) and because the repeated-part variations (meristic variations) that especially interested him could be understood as arising from the same mechanical forces—‘the physics of Division’—that determined species types (pp. 70–71, quotation on p. 70).
45 Bateson (1913a), pp. 268-269; Bateson (1913b), pp. 69-70 (the phrase ‘meristic forces’ is on p. 70). A distinction between the meristic/physical and substantive/chemical goes back to Materials (1894, pp. 70-73). Darden (1977, esp. p. 98) also noted the centrality and singularity of Bateson’s notion of segregation, and the continuity with his earlier work.
46 In Bateson’s view, cases of, for instance, recessive-character fruits growing on dominant-character trees oblige us ‘to admit that it is not solely the reduction-divisions which have the power of effecting segregation’; Bateson (1913a), pp. 272-274, quotation on p. 273. On heredity as like symmetry and variation as like asymmetry, see Bateson (1913a), pp. 274-278—another idea that goes back to Materials (1894, p. 34). On the range of Bateson’s research projects at the John Innes Institution to do with segregation beyond meiosis, see Olby (1989b), p. 508.
47 Mendel (1866), p. 41. It should be noted that other translators have rendered the passage differently. Gliboff, for example, gives ‘numerical’ instead of ‘statistical’; see Gliboff (1999), p. 226. Certainly that chimes better with Ted Porter’s judgement that Mendel’s reasoning about heredity was not strictly speaking statistical (in contrast with Galton’s); see Porter (1986), p. 135.
48 See Kragh (2002), esp. p. 45 on how, when physicists wrote on vortex atoms, ‘only rarely did they connect their ideas with measurable properties of matter’.
49 On the Fosterian education that Bateson’s cohort at Cambridge received, see Geison (1978), pp. 118–119; on the role of measurement, apparatus and experiment in that education, see p. 164.
50 On the early period of Bateson’s discipleship, see, e.g., Gillham (2001), pp. 288–292; see p. 261 for Galton’s polygon (reproduced from Galton (1889), p. 27). For Bateson’s enthusiasm in Materials for Galton on ‘Organic Stability’, see Bateson (1894), p. 36. On the intellectual adjustments that Bateson the Galtonian made after encountering Mendel’s work, see Olby (1987). On Galton’s support for Bateson’s experimental studies of heredity in the 1890s, see Schwartz (2008), pp. 60–61.
51 Bateson (1917), p. 209.
52 Coleman (1970), p. 265.
53 I have not except in an incidental way dealt here with the political values of precision; but, a propos Coleman’s influential claim that Batesonian Mendelism was politically conservative, consider Punnett’s little book on the science. ‘The most interesting part of the Mendelian theory is that it is a mathematical one’, wrote Gaylord Wilshire, in his preface as publisher of the American edition of Mendelism, ‘and this is what charms me regarding the theory of mutation in society. It too is a mathematical one. You can count up the number of machines and count up the number of men, and can prophesy the time almost exactly when Socialism must come in order to make a balance between production by machines and consumption by men.’ With Mendelism as with others sciences, the control that precision promised—and sometimes yielded—had appeal across the political spectrum. (Other books in Wilshire’s list included Enrico Ferri’s Socialism and modern science (Darwin, Spencer, Marx) and Marx’s A contribution to the critique of political economy.) Punnett (1909), pp. 5–6, back matter. On precision’s role in ‘extending control’, see Wise (1995), pp. 4–6.
54 See Hodge (1992), esp. pp. 238–262; also Fisher (1930) and Depew and Weber (1995), ch. 10. For a lucid exposition of the eugenical parts of the 1930 book, see Gould (1996), esp. p. 306 on Fisher’s reliance on Galton’s ‘infertile heiress’ argument.
55 Punnett (1909), p. 62. Cf. Bateson (1902), p. 35, on Mendel’s experiments as ‘worthy to rank with those which laid the foundations of the Atomic laws of Chemistry’. See also Keller (2000), pp. 18–20.
56 On the undergraduate Fisher as a reader of the 1909 edition of Bateson’s Mendel’s principles of heredity, see Bennett (1983), pp. 7–8.
57 Two books more or less define the historiographic status quo on the controversies: Provine (2001) and Kim (1994); but for longer perspectives, see also Bowler (1989) and Schwartz (2008). On how much Mendelism owed to the early criticism, see Olby (1994). For part of the story of how Mendelians immunized their science against the charges that Weldon levelled, see Charnley & Radick (2009) and, in less abbreviated form, Charnley and Radick (forthcoming). For more general reflections on the Weldonian biology that might have been, see Radick (2005).
58 On the evolutionary biologist Ernst Mayr’s role in popularizing biology’s separateness from physics, see Milam (2010), esp. 312.
59 Schaffer (forthcoming).
60 Poon, McLeish, & Donald (2002), quotation on p. 25.

References

Bateson, B.  (1984).  William Bateson, F.R.S., naturalist: His essays & addresses, together with a short account of his life.  New York & London. Facsimile reprint of the 1928 edition.

Bateson, W.  (1894).  Materials for the study of variation treated with especial regard to discontinuity in the origin of species.  London: Macmillan.

Bateson, W.  (1900).  Problems of heredity as a subject for horticultural investigation. In B. Bateson (1984) (pp. 171-180). 

Bateson, W.  (1902).  Mendel’s principles of heredity: A defence.  With a translation of Mendel’s original papers on hybridisation.  Cambridge: Cambridge University Press.

Bateson, W.  (1913a).  Mendel’s principles of heredity.  3rd impression of 1909 edition with additions.  Cambridge: Cambridge University Press.

Bateson, W.  (1913b).  Problems of genetics.  New Haven: Yale University Press.

Bateson, W.  (1917).  Gamete and zygote.  A lay discourse.  In B. Bateson (1984), pp. 201-214.

Bateson, W.  (1924).  Progress in biology.  In B. Bateson (1984) (pp. 399-408).

Baxter, A., & Farley, J.  (1979).  Mendel and meiosis.  Journal of the History of Biology, 12, 137-173.

Bennett, J. H.  (Ed.)  (1983).  Natural selection, heredity, and eugenics: Including selected correspondence of R. A. Fisher with Leonard Darwin and others.  Oxford: Clarendon Press.

Berry, A.  (2001).  Wrinkled v. round.  London Review of Books,8 February, 28-29.

Bowler, P. J.  (1989).  The Mendelian revolution: The emergence of hereditarian concepts in modern science and society.  Baltimore: Johns Hopkins University Press.

Brock, W. H.  (1992).  The Fontana history of chemistry.  London: Fontana Press.

Carter, B., & Carter, M. S.  (2002).  Latitude: How American astronomers solved the mystery of variation.  Anapolis, MD: Naval Institute Press.

Charnley, B., & Radick, G.  (2009).  Plant breeding and intellectual property before and after the rise of Mendelism: The case of Britain.  In J-P. Gaudillière, D. J. Kevles, & H.-J. Rheinberger (Eds.), Living properties: Making knowledge and controlling ownership in the history of biology (pp. 51-56).  Preprint 382.  Berlin: Max Planck Institute for the History of Science.

Cock, A. G.  (1983).  William Bateson’s rejection and eventual acceptance of chromosome theory.  Annals of science, 40, 19-59.

Cock, A. G., & Forsdyke, D. R.  (2008).  Treasure your exceptions: The science and life of William Bateson.  New York: Springer.

Coleman, W.  (1970).  Bateson and chromosomes: Conservative thought in science. Centaurus, 15, 228-314.

Corcos, J. F., & Monaghan, F. V.  (1993).  Gregor Mendel’s Experiments on plant hybrids: A guided study.  New Brunswick, NJ: Rutgers University Press.

Correns, C.  (1900).  G. Mendel’s law concerning the behavior of the progeny of varietal hybrids.  Trans. in C. Stern, & E. R. Sherwood (Eds.), The origin of genetics: A Mendel source book (pp. 119-132).  San Francisco & London: W. H. Freeman, 1966. 

Cuvier, G., & Latreille, P. A.  (1831).  The animal kingdom arranged in conformity with its organization.  Vol. 1.  New York: G. & C. & H. Carvill.

Darden, L.  (1977).  William Bateson and the promise of Mendelism.  Journal of the History of Biology, 10, 87-106.

Depew, D. J., and Weber, B. H.  (1995).  Darwinism evolving: Systems dynamics and the genealogy of natural selection.  Cambridge, MA and London: MIT Press.

Dettelbach, M.  (1996).  Humboldtian science.  In N. Jardine, J. A. Secord, & E. C. Spary (Eds.), Cultures of natural history (pp. 287-304).  Cambridge: Cambridge University Press.

Fisher, R.  (1930).  The genetical theory of natural selection.  Oxford: Clarendon Press.

Franklin, A., Edwards, A. W. F., Fairbanks, D. J., Hartl, D. L., & Seidenfeld, T. (2008).  Ending the Mendel-Fisher controversy.  Pittsburgh: University of Pittsburgh Press.

Galton, F.  (1872).  On blood-relationship.  Proceedings of the Royal Society of London, 20, 394-402.

Galton, F.  (1875).  A theory of heredity.  Contemporary Review, 27, 80-95.

Galton, F.  (1876).  A theory of heredity.  Journal of the Anthropological Institute, 5, 329-348.

Galton, F.  (1884).  Free-will – observations and inferences.  Mind, 9, 406-413.

Galton, F.  (1904).  Eugenics: Its definition, scope and aims.  American Journal of Sociology, 10, 1-6.

Galton, F.  (1908).  Memories of my life.  London: Methuen.

Gayon, J.  (1998).  Darwinism’s struggle for survival: Heredity and the hypothesis of natural selection.  Trans. M. Cobb.  Cambridge: Cambridge University Press. 

Geison, G. L.  (1978).  Michael Foster and the Cambridge School of Physiology: The scientific enterprise in late Victorian society.  Princeton: Princeton University Press.

Gillham, N. W.  (2001).  A life of Sir Francis Galton: From African exploration to the birth of eugenics.  Oxford: Oxford University Press.

Gliboff, S.  (1998).  Evolution, revolution, and reform in Vienna: Franz Unger’s ideas on descent and their post-1848 reception.  Journal of the History of Biology, 31, 179-209.

Gliboff, S.  (1999).  Gregor Mendel and the laws of evolution.  History of Science, 37, 217-235.

Gould, S. J.  (1996).  The smoking gun of eugenics.  In Dinosaur in a haystack: Reflections in natural history (pp. 296-308).  London: Jonathan Cape.

Hacking, I.  (1990).  The taming of chance.  Cambridge: Cambridge University Press.

Harman, P.  (1998).  The natural philosophy of James Clerk Maxwell.  Cambridge: Cambridge University Press.

Harman, P.  (Ed.)  (2002).  The scientific letters and papers of James Clerk Maxwell, vol. 3.  Cambridge: Cambridge University Press. 

Hodge, M. J. S.  (1992).  Biology and philosophy (including ideology): A study of Fisher and Wright.  In S. Sarkar (Ed.), The founders of evolutionary genetics: A centenary reappraisal (pp. 231-293).  Dordrecht: Kluwer.  Reprinted in M. J. S. Hodge, Before and after Darwin: Origins, species, cosmogonies, and ontologies (ch. 13).  Aldershot: Ashgate, 2008.    

Keller, E. F.  (2000).  The century of the gene.  Cambridge, MA & London: Harvard University Press.

Kim, K.-M.  (1994).  Explaining scientific consensus: The case of Mendelian genetics.  New York and London: Guilford Press. 

Kragh, H.  (2002).  The vortex atom: A Victorian theory of everything.  Centaurus, 44, 32-114.

Livio, M.  (2002).  The golden ratio: The story of phi, the world’s most astonishing number.  New York: Broadway Books.

Maxwell, J. C.  (1873).  Essay for the Eranus Club on science and free will.  In P. M. Harman (Ed.), The scientific letters and papers of James Clerk Maxwell, vol. 2 (pp. 814-823).  Cambridge: Cambridge University Press.      

Maxwell, J. C.  (1875).  Atom.  Reprinted in facsimile in E. Garber, S. G. Brush, & C. W. F. Everitt (Eds.), Maxwell on molecules and gases (pp. 176-215).  Cambridge, MA and London: MIT Press, 1986.

Mendel, G.  (1866).  Experiments in plant-hybridisation.  Trans. in Bateson (1902), pp. 40-95.

Merz, J. T.  (1965).  A history of European scientific thought in the nineteenth century.  Vol. 2.  New York: Dover.  First published in 1903.

Milam, E. L.  (Forthcoming.)  The equally wonderful field: Ernst Mayr and  organismal biology in the 1960s.   

Montgomery, W. M.  (1970). The origins of the spiral theory of phyllotaxis.  Journal of the History of Biology, 3, 299-323.

Newman, S. A.  (2007).  William Bateson’s physicalist ideas.  In M. D. Laubichler, & J. Maienschein (Eds.), From embryology to evo-devo: A history of developmental evolution (pp. 83-107).  Cambridge, MA and London: MIT Press.

Olby, R. C.  (1985).  Origins of Mendelism.  2nd edition.  Chicago & London: University of Chicago Press.  1st edition in 1966.

Olby, R. C.  (1987).  William Bateson’s introduction of Mendelism to England: A reassessment.  British Journal for the History of Science, 20, 399-420.

Olby, R. C.  (1989a).  The dimensions of scientific controversy: The biometric-Mendelisn debate.  British Journal for the History of Science, 22, 299-320.

Olby, R. C. (1989b).  Scientists and bureaucrats in the establishment of the John Innes Horticultural Institution under William Bateson.  Annals of Science, 46, 497-510.

Olby, R. C.  (1994).  Sociology of science, sociology of knowledge, and scientific consensus.  In K.-M. Kim, Explaining scientific consensus: The case of Mendelian genetics (pp. 191-201).  New York and London: Guilford Press. 

Pearson, K.  (1906).  Walter Frank Raphael Weldon.  1860-1906.  Biometrika, 5, 1-52.

Pearson, K.  (1908).  On a mathematical theory of determinantal inheritance, from suggestions and notes of the late W. F. R. Weldon.  Biometrika, 6, 80-93.

Pence, C.  (Forthcoming).  ‘Describing our whole experience’: The statistical philosophy of W. F. R. Weldon. Poon, W., McLeish, T., & Donald, A.  (2002).  Soft condensed matter physics: where physics meets biology.  Physics education, 37, 25-33. 

Porter, T. M.  (1986).  The rise of statistical thinking 1820-1900.  Princeton: Princeton University Press.

Provine, W. B.  (2001).  The origins of theoretical population genetics.  Chicago: University of Chicago Press.  First published in 1971.

Punnett, R. C.  (1909).  Mendelism.  American reprint of the 2nd edition of 1907 (1st edition in 1905).  New York: Wilshire. 

Radick, G.  (1998).  The Origin unbound.  Studies in History and Philosophy of Biological and Biomedical Sciences, 29C, 349-357.

Radick, G.  (2005).  Other histories, other biologies.  In A. O’Hear (Ed.), Philosophy, biology and life (pp. 21-47).  Cambridge: Cambridge University Press.

Renwick, C.  (In press)  From political economy to sociology: Francis Galton and the social-scientific origins of eugenics.  British Journal for the History of Science

Richards, R. J.  (2008).  The tragic sense of life: Ernst Haeckel and the struggle over evolutionary thought.  Chicago & London: University of Chicago Press.

Schaffer, S.  (1999).  Entranced by the factory.  London Review of Books, 29 April: 31-32.

Schaffer, S.  (2000).  Who’s minding the garden?  In A. Lowe, & S. Schaffer (Eds.), N01SE (exhibit catalogue).  Cambridge: Whipple Museum; London: Wellcome Institute.

Schaffer, S.  (Forthcoming).  How disciplines look.

Schwartz, J.  (2008).  In pursuit of the gene: From Darwin to DNA.  Cambridge, MA. & London: Harvard University Press.

Turner, H. H.  (1904).  Astronomical discovery.  London: Edward Arnold.

Weldon, W. F. R.  (1902a).  Mendel’s laws of alternative inheritance in peas.  Biometrika, 1, 228-254.

Weldon, W. F. R.  (1902b).  On the ambiguity of Mendel’s categories.  Biometrika, 2, 44-55.

Weldon, W. F. R.  (1905).  Inheritance in animals and plants. In T. B. Strong (Ed.), Lectures on the Method of Science (pp. 81-109).  Oxford: Clarendon Press.

Wise, M. N. (Ed.) (1995).  The values of precision.  Princeton: Princeton University Press.