THE FRENCH SCHOOL OF GENETICS: From Physiological and Population Genetics to Regulatory Molecular Genetics.

RM Burian [1]

J Gayon [2]

Key Words history of genetics, lysogeny, population genetics, physiological genetics, regulatory genetics

* Abstract French genetics had unusual beginnings. There are clear indications that the French biological establishment resisted Mendelian genetics strenuously from about 1910 to 1940. From about 1930 to 1950 several unconventional research programs with a strongly physiological orientation paved the way for the full entrance of French biology into genetics after World War II. This review examines some salient features of this history to clarify the strengths, weaknesses, and distinctive features of French genetics until about 1965. We suggest that after that date French genetics slowly merged into the international mainstream as genetics has become a largely molecular discipline.

INTRODUCTION

In this review we explore some aspects of French biology before World War II to highlight certain key features of the founding of regulatory genetics in France in the 1950s and 1960s The history we cover is puzzling: For nearly half a century mainstream French biologists expressed extreme misgivings about Mendelian and chromosomal genetics and virtually ignored it in university curricula. Yet in the 15 years after World War II they did much groundbreaking work in cytoplasmic, physiological, and regulatory genetics, culminating in the work that won the Nobel Prize for Jacob, Lwoff, and Monod for their contributions to regulatory genetics in microorganisms. To understand these rapid and dramatic changes, we examine the period from 1900 (the rediscovery of Mendel's laws) to 1965 (the awarding of the Nobel Prize, which marks the return of French genetics to the international mainstream). In our interpretation, the emergence of a French school of molecular genetics in the 1950s was not simply a decisive victory of a camp of enlightened scientists over obscurantist and retrograde colleagues. Rather, it also represented the culminating success of little-appreciated, autonomous traditions of work on heredity developed in independence of the standard Mendelian-chromosomal genetics that became the international mainstream during the first half of the century.

In what follows, we describe the reception and rejection of Mendelian genetics in the first three decades of the century. From 1930 to 1950, there was some interest in genetics, but several unorthodox lines of research paved the way, unforeseeably, for the achievements of the 1950s. The last part of the paper deals with those achievements, focussing on physiological genetics, cytoplasmic heredity, and nucleo-cytoplasmic relations. We touch on institutional as well as conceptual aspects of the history and locate some developments in an international context.

FRENCH BIOLOGISTS AND GENETICS, 1900--1930

To understand the resistance of French biologists to Mendelian genetics, it is important to discuss three matters: the diffusion of the new discipline within the scientific community, the teaching of genetics, and research directed toward problems of heredity. Each of these aspects tells a different story [25].

Scholarly Diffusion

We have studied six widely read French biological journals from 1900 to 1930. [1] Mendelian genetics was thoroughly discussed from the very beginning. The very first publication of the rediscovery, published by De Vries in March 1900, in the Comptes Rendus de l'Academie des Sciences [49], presented the law of disjunction without mentioning Mendel. This paper was reviewed twice in l'Annee Biologique for 1900: by Cuenot [36] and by the editor, Yves Delage, who devoted a full page to it in his introductory overview of the volume. Between 1901 and 1903, l'Annee Biologique published abstracts of major papers on Mendelism by Bateson, Castle, Correns, Cuenot, Darbishire, De Vries, Doncaster, Haecker, Tschermak, and Wilson. The 1902 volume included an extensive review of Mendelian research by Cuenot [38]. The journal continued in this vein through 1914. Again, Cuenot, the leading French contributor to the early development of the new discipline, published Mendelian papers in five of the six journals on our list and w as awarded the Cuvier Prize of the Academie des Sciences in 1911, with special mention of his genetic experiments. Those experiments stopped, however, in 1914.

Anti-Mendelians also reported Mendelian research fully, although critically. Thus, the Bulletin Scientifique de la France et de la Belgique, a bastion of neo-Lamarckism, published (and criticized) numerous accounts of Mendelian work, including a translation of Mendel's 1865 memoir in 1907. We must also mention the Fourth International Conference of Genetics, held in Paris in 1911 (154a). The Conference was organized at the instigation of the vilmorin family [81], proprietors of a major seed company, and was attended by an impressive number of French biologists, though some were critics of genetics. In short, even though Mendelism was not widely accepted, it was widely known and discussed before World War I.

Textbooks and Teaching

The failure of genetics to enter into the curricula of French universities in the first decades of this century is enormously important, for it shaped the methodological and conceptual matrix of the next generations of scientists. International comparisons are useful here. By the onset of World War I, there were chairs of genetics (sometimes with different names) in England (Punnett), Germany (E Baur), and the United States (EB Brown). In Moscow, A Serebrovsky accepted a chair of genetics in 1930. In France, the first chair of genetics was created for Felicien Boeuf in 1936 at the National Institute of Agronomy in Paris [153], but this was not a University chair and was not widely influential. The first university chair in genetics was created only in 1946 for Boris Ephrussi.

Textbooks also provide striking evidence. At least 13 English or American treatises or textbooks were specifically devoted to genetics before 1920, many of them in multiple editions. Five comparable books were published in German [listed in [25]]. No synthetic book on genetics was published in France until the mid-1920s. To be sure, Cuenot included significant chapters on Mendelian heredity in the successive editions [1911, 1922, 1932] of his book on the Genesis of Animal Species [43]. Still, the first systematic book on Mendelian genetics in French was a poorly distributed translation of Morgan, Sturtevant, Muller & Bridges' Mechanism of Mendelian Inheritance, published in Belgium in 1923. The first French textbook was published a year later by Emile Guyenot [81]; this remarkable book greatly influenced the young biologists of the time. Its successive editions [1930, 1942, 1948] remained the major French-language source of information into the 1950s for anyone seeking serious initiation into genetics.

Finally, genetics was barely taught in colleges and universities until after 1945. The only exception before 1930 was Nancy; there, Cuenot included genetics in his zoology courses from before World War I. In Paris in the 1920s, Blaringhem, a neo-Lamarckian, introduced some genetics at the Ecole Normale Superieure [10], as did Caullery in the Sorbonne, in the form of a 12-hour optional course [32]. Also in the 1920s, a course entitled Genetique, phytotechnie and botanique appliquee" was taught at the National Institute of Agronomy (153, p. 109). But there was nothing more in the first three decades of the century; it was only in 1948 that the first "Certificate of Genetics" was instituted in university curricula under Ephrussi's supervision. The contrast with other major scientific nations is dramatic.

Research

The story of genetic research during these decades is even stranger. Thanks to Lucien Cuenot [1866-1951], the French were well represented in the first wave of Mendelian research. But, after a brilliant start, Mendelian genetics was eliminated from mainstream biological research in France in the second and third decades of twentieth century. Cuenot, briefly a Darwinian selectionist [106], switched, shortly before 1900, to a more saltational view of evolution that he eventually elaborated under the name of "the theory of preadaptation" [76]. From 1900-1910, Cuenot was one of the most productive geneticists. His contributions include (a) extension of the laws of disjunction and independent assortment to the animal kingdom, specifically mice [37]; (b) the discovery of multiple alleles [40,41]; (c) recognition of interaction among different Mendelian factors [41]; (d) recognition that a given Mendelian determinant can mask the effect of other Mendelian factors (epistasis in a more modem sense) [41]; (e) recogniti on of lethal homozygotes [42]; (f) the first statement of the hypothesis that gene function is related to production of enzymes [39]; and (g) pioneering work on the genetics of cancer, especially the recognition that tissues with a given genotype can behave differently according to the genotype of surrounding tissues [[45] and seven later papers]. Although he completely abandoned experimental research in genetics after 1914, Cuenot's reputation in mammalian genetics remained strong: In 1928, he reviewed the genetics of mice for Bibliographia Genetica [44]. His stocks of mice were destroyed in World War I. When he returned to Nancy in 1918, it was clear that the Morgan school had already made the big breakthrough. Cuenot, "who wanted to be the first" (JG interview with Andree Tetry), gave up. He also dissuaded his students from writing PhD dissertations in genetics because he thought that they would not find positions in France.

This last point reflects the French resistance to genetics. France was proud of Cuenot, but in general there was strong intellectual reluctance to pursue Mendelism, as is illustrated by the treatment of Emile Guyenot. In 1909, he began a PhD dissertation under the direction of Maurice Caullery. The theoretical objective (strongly supported by a major neo-Lamarckian, Etienne Rabaud) was to show that Morgan's Mendelian results were artefacts caused by improper control of the experimental conditions. The idea was that some "mutations" were conditions caused by failure to control the nutrition of the flies or by infections stemming from failure to establish a rigorously aseptic experimental environment. By 1917, in spite of the war, Guyenot had counted 400,000 flies and examined a large number of mutants. But the mutants did not change in altered environments; they arose in constant environments, and Mendel's laws were verified whatever the environment. Guyenot's dissertation, published by Rabaud in the Bulletin Biologique de la France et de la Belgique (80a), led to a violent dispute between them solely because Guyenot had supported Mendelism. In the end, Guyenot did not find a position in France. He was recruited to Geneva, where he acquired considerable influence as a geneticist, especially through his 1924 textbook. But he had no students in France, and, by 1930, no one there was carrying out fundamental genetic research.

Explanation of the Delayed Development of Genetics in France: A General Scheme

The resistance of the French biological community to genetics stems from at least four causes. First, certain intellectual traditions contributed to the reluctance to accept Mendelian genetics. A number of these that were influential near the turn of the century are of particular concern here. Consider, first, the influence of resistance to Darwinian evolution, quite virulent in France. Around 1900, French paleontologists, among the few who had been sympathetic to some aspects of Darwinism and who had expended considerable effort in constructing phylogenetic trees in the nineteenth century, abandoned this practice on grounds that no demonstration of filiation between species was feasible. Instead, the standard paleontological view became that one could only record the relative prevalence of different forms of organisms in successive strata, yielding results that simply could not reveal the process or sequence by which a species was formed [77a]. Meanwhile, under positivist influences, many biologists, includi ng many who accepted evolution, became increasingly skeptical about the possibility of identifying causes of transformation of organismal properties and transmission of particular traits across a series of generations. They argued that neither selection nor modes of transmission were accessible. Again, adherents to the Bernardian ideal for biological sciences rejected Mendelian and chromosomal genetics for its formal and "unphysiological" character. A related argument was that Mendelians gave the nucleus exorbitant powers and neglected its physiological interactions with the cytoplasm so that no sense could be made of how Mendelian factors act. More generally, many biologists influenced by French positivism denounced the "metaphysical" character of Mendelian "factors"--speculative hypothetical entities in the service of a discredited preformationist approach to biological problems [25]. These distinct traditions linked up to reinforce a generalized skepticism about the ability of biologists to determine the c auses of evolution or of inherited traits and their transmission; the most that could be scientifically ascertained was a description of the "before and after" for changes of interest.

A second source of resistance concerns the connection of genetics to eugenics. Like it or not, the countries where genetics, which favored "hard heredity" and denied the inheritance of acquired characters, flourished from 1900 to 1930 (England, USA, Germany, the Scandinavian countries), are precisely the ones that developed strong eugenic traditions and legislation. Although a French Society of Eugenics was created in 1912 (with Cuenot as one of its most active members), it never achieved influential prominence. Historians connect this to the French obsession with their demographic decline vis-a-vis Germany--French eugenics, such as it was, took the paradoxical form of providing the best conditions for producing a maximum of mothers and babies [31,53,74,77, 142].

A third is the failure of French academic biology to establish solid interactions with agricultural research. The countries where Mendelism flourished all had forged close ties between agricultural engineering and experimental biology (again, the United States, England, Germany, the Scandinavian countries, and perhaps, ambiguously, Russia). In France, the Vilmorin Company tried hard to develop such connections--genetics was a crucial discipline for its survival as the foremost seed company in the world before 1914 [79]. But, in practice, the worlds of university research and plant and animal breeding remained widely separated until the creation of the Institut National de Recherche Agronomique in 1921.

Finally, there is the sheer loss of a generation of young men in World War I. So few students returned from the war that many disciplines were very seriously underdeveloped. In the climate we have described, genetics did not rank high on the list of the returnees' priorities. The lack of career paths and the discouraging climate regarding genetic investigations kept those biologists who sought work within the French mainstream away from that discipline.

None of these causes was sufficient, by itself, to account for the delayed development of genetics. Their conjunction, however, provides a reasonable account of that delay.

THE ROUTES TO GENETICS, 1930-1950

There are signs of a significant change in some French biologists' attitude toward genetics in the l930s. Mention of the topic in a few modest and nonmandatory introductory courses in the 1920s had produced some effect on the new generation. Jean Rostand's popular book of 1930, From Flies to Humans [141], had a considerable impact, and Guyenot's textbook of 1924 [81] served as a major point of entry for those who wanted to get reliable information on genetics in French.

In the 1930s, a handful of young biologists embarked on innovative research programs that defined the specific routes that led French biology into the international genetics community. None of them were formally trained in genetics in their courses or in French laboratories (how could it be otherwise?). They all worked outside the university system and had few or no teaching obligations. Georges Teissier and Philippe L'Heritier, originally mathematicians working at the Ecole Normale Superieure, were the only ones to have some teaching duties. Boris Ephrussi worked at the Institut de Biologie Physico-Chimique (or Rothschild Foundation), a research institute founded in 1926 to foster innovative experimental research in biology, while Andre Lwoff, Eugene Wollman, and later Jacques Monod worked at the Pasteur Institute. These people knew each other, often spent summers together at the marine biological station in Roscoff, and published together in all possible combinations. They were all involved in internationa l networks of some sort or other. Four of them (Ephrussi, L'Heritier Monod, and Lwoff) received Rockefeller Foundation support in the 1930s. The first three used this support to work in US laboratories, Lwoff in Germany and England. Finally, considered in terms of genetics, all their research programs were either marginal or unconventional, but all…