James Meadows Rendel 1915-2001

Written by Ian Franklin, Geoff Grigg and Oliver Mayo.

Life

James Meadows Rendel was born on 16 May 1915 in England. He moved to Australia in 1951 to join CSIRO and was appointed Chief of the Division of Animal Genetics in 1959. He was elected to the Australian Academy of Science in 1960, retired from CSIRO in 1980 and died on 4 February 2001. His influence on genetics and the development of the theory and practice of animal breeding in Australia was profound.

After his election to the Academy, Rendel served on its Council in 1963-5 and 1971-4, being a vice-president in 1973-4. He was Burnet Lecturer in 1981.

Rendel's family background (important to understanding an Englishman) was intellectual. Although his father, Colonel Richard Meadows Rendel, was a professional soldier, there were connections to the Bloomsbury group on both sides: his father's uncle was Lytton Strachey, and his mother's sister was the diarist Frances Partridge. Rendel's first marriage, on 22 October 1938, was to the poet Joan Adeney Easdale, a protégée of Leonard and Virginia Woolf who published some of her poems in their Hogarth Press. Friends included Professor Lionel Penrose the human geneticist (whose brother was a champion of cubism and cubists in England) and the Mitchison clan. As a student, Rendel played the flute, was fascinated by ballet and kept a chameleon. He received part of the normal education of an upper- middle-class Englishman of his time (Rugby School), but went to University College London rather than Oxbridge and completed his PhD as a student of J. B. S. Haldane, the great geneticist and left-wing science popularizer. Working with Haldane gave him a breadth as a scientist that was a great boon throughout his career, but also left him in due course with a permanent physical disability. We believe that Rendel and John Maynard Smith, another notable evolutionary geneticist, were Haldane's only PhD students, something that seems unlikely and which we have been unable to confirm.

During the Second World War, when Rendel was attached to RAF Coastal Command, he took part in Haldane's experiments on escape from submarines (critical after a peace-time disaster). Haldane never spared himself as a guinea-pig, and those of his colleagues with as much courage took the same risks. One experiment involving a high compression chamber was disastrous and almost fatal for Rendel, as both of his lungs were punctured and he was left with permanent pulmonary problems. General details of this work can be found in Ronald Clark's biography of Haldane (Clark 1968), but it is characteristic of Rendel, who helped Clark considerably with the book, that the reference to this accident does not mention Rendel by name. As most of the actors in these events are no longer alive, Clark's book remains the best published source, incomplete though it is.

After the war, Rendel moved to Edinburgh to join the legendary animal genetics research group put together in the University of Edinburgh by C. H. Waddington, a man of such huge ego that he never hesitated to appoint better scientists than himself to his Institute. Two such were Jim Rendel and Alan Robertson, both of whom were known to Waddington from his time as Chief of Operational Research in Coastal Command. (They had worked on methods for detecting German U-boats at sea.) They were put in charge of a dairy research programme, and together they made several vital advances in the design and analysis of dairy breeding programmes, establishing principles that laid the foundation for the widespread use of artificial insemination in dairy progeny testing programmes. Later, Rendel implemented these ideas in CSIRO's tropical dairy breeding initiative. Later still, Robertson (1977) wrote a biographical memoir of Waddington that stresses canalisation and yet, perhaps at first sight surprisingly, does not mention Rendel. However, Robertson did not agree with Rendel's approach to Drosophila bristle epigenetics (of which much more below), and to mention his old friend would have been to criticize him.

Rendel was then recruited from Edinburgh in 1951 by Sir Ian Clunies Ross, Chairman of CSIRO, to set up a CSIRO group at the University of Sydney to teach animal genetics and to develop and supervise a programme of research into animal breeding methods encompassing the domestic fowl, sheep, dairy and beef cattle. Rendel's recruitment, along with those of Otto Frankel (CSIRO, Canberra) and David Catcheside (University of Adelaide), was the culmination of an intensive effort by Clunies Ross to re-establish the science of genetics in Australia following its decline in the pre- and post-war years (see McCann and Batterham 1993)

When he arrived from Edinburgh, Rendel and the two colleagues he brought with him were housed in premises at the University of Sydney where they began to teach genetics in both the Science and Veterinary Science Faculties. As he developed his Section, it grew in size and responsibilities and CSIRO made it an independent Division, of Animal Genetics, in 1959.

His son Sandy Rendel notes that 'Jim really liked the farmers he came in contact with. On the boat out (the P & O liner Oronsay) he and my mother became friends with Charles and Amy Cooper. Charlie Cooper owned Kunanadgee, 3000 acres on the River Murray at Corowa. It was one of the farms Jim visited on his familiarisation tour after arriving in Australia. Subsequently it was involved in the release of myxo.[1] I spent a lot of my school holidays there. Jim also enjoyed his trips to the research station at Cunnamulla and working with Bill Gunn and the meat board' (S. Rendel, pers. comm., 2004)

Whilst he enjoyed his involvement in animal breeding – and cattle breeding was a passion – Rendel's primary interest was in how genes worked in the animal. In his first year in Australia, he joined with a number of other geneticists to found the Genetics Society of Australia. He set up groups in the Section of Animal Genetics to study the fundamentals of genetics, using mice, Drosophila and Paramecium as experimental organisms. In the 1960s, he realised the future impact that molecular genetics must have in animal improvement, and set out to establish molecular biology in the Division. This group later was spun off as CSIRO's Molecular and Cellular Biology Unit. He had a particular personal interest in the genetics of developmental processes in the whole animal (here the influence of Waddington during his five years in Edinburgh is evident). Rendel explored and wrote about novel ideas on developmental 'canalisation', discussed in detail in the next section; his only published book, Canalisation and Gene Control, covers the topic.

Although Rendel continued his own research with Drosophila, and built new research programmes in molecular and developmental biology within his own Division, he recognised his responsibility to provide 'something for now, something for later' (in Justus von Liebig's words) very broadly, and he had the vision of, and the resources to realise that vision in, new breeds as well as new scientific concepts.

In the application of genetics to animal breeding, Rendel's main personal contribution was to the development of new breeding programmes, especially for northern Australia. CSIRO had already imported Indian and African cattle to develop a livestock industry for tropical and sub- tropical regions, and Rendel encouraged basic research into understanding the interactions between adaptation to stressful environments and productivity in those environments. The improvement of adaptation in beef cattle by use of African germplasm resulted in the Belmont Red; in dairy cattle, Indian cattle and European dairy cattle were used to produce the Australian Milking Zebu (AMZ). Had typical Australian parochialism not led to a rival programme in Queensland that produced a competing breed, the Australian Friesian Sahiwal (AFS), Australia might have led the tropical world in developing a sustainable dairy breed and production system. In contrast, the beef cattle industry in Australia's north has been a major success. In poultry, Rendel's ideas on the practical application of canalisation theory to decrease the inter-egg interval in layers made, through Bruce Sheldon, a major contribution to the egg industry before that industry closed genetics research in favour of importing germplasm.

A new tropical beef cattle research laboratory was built in Rockhampton at about the time Rendel retired as Chief of Division, and it was named the James Rendel Laboratory to recognise his contribution. There is no comparable memorial to his contribution to sheep breeding, where, with his quiet support and critical direction, Helen Newton Turner and others pioneered objective measurement of fleece traits and struggled to help sheep breeders, in many cases against their will, to apply the same successful science to the wool industry as had revolutionized pigs, poultry and dairy.

He knew the importance of encouraging young scientists and was always generous with his time and CSIRO's resources to help those whom he judged deserving. Stuart Barker, Jim Peacock and the authors of this memoir are among the many who benefited from his quiet early encouragement. Under his leadership, the Animal Genetics section was egalitarian and a fount of ideas that reverberated through the emerging Genetics community. His arguments were always stimulating but often obscure, especially to those who saw animal breeding as a science rooted in statistics rather than biology.

Upon his retirement from CSIRO in 1980, Rendel briefly took up a visiting professorship at Harvard to collaborate with his old friend Richard Lewontin. Subsequently he and his wife Tresham moved to live in a fine house on a smallholding at Drinkstone Green in Suffolk, England, where he bred Booroola sheep which he had imported from Australia. This was following in his father's footsteps, for Colonel Rendel had retired early to raise poultry in Kent, returning to active service for the Second World War.

Australia, however, continued to beckon and seven years later he and Tresham returned to Sydney and to Wentworth Falls in the Blue Mountains above Sydney where they settled. An old friend and colleague, Bill Sobey, writes that in his opinion Rendel did not 'ever become an Australian', he was too English, but he had a deep affection for his adopted land.

In this second retirement, Rendel maintained his lifelong interest in thinking and writing about science, and wrote an entertaining and challenging book about the role of common sense in judging science.

He was survived by his second wife Tresham (Marie Tresham Davies; Tresham was a family name, one of her ancestors being Francis Tresham, a co-conspirator with Guy Fawkes in the English Gunpowder Plot of 1605), his six children (Jane Susan Robertson, Polly Mary Virginia Woods, Alexander Meadows Rendel, twins Julia Margaret Szulerowski and Richard James Rendel and Francis Kate Hayes), ten grandchildren and one great-grandchild.

Work

Early work and influences

Rendel's earliest work, as a PhD student in London, was a study of the relationship between egg size and hatchability in ducks. He published a number of papers arising from this work, demonstrating in particular that intermediate egg weights are favoured by natural selection. Undoubtedly, this work drew his attention to mechanisms that allow for reduced variability, a topic he revisited later in his Drosophila experiments. Later, when he joined Waddington's group in Edinburgh, he was drawn to the application of genetics to animal breeding and formed a close association with Alan Robertson. Together, Rendel and Robertson formulated a set of protocols for dairy improvement based on progeny testing and the widespread use of artificial insemination. This work was seminal in the design and analysis of dairy breeding programmes, especially in Europe, and their methodologies were in many ways superior to those being developed at the same time by Hazel and Lush in the USA. Their work led to a decline in the use of testing stations and the widespread use of family selection in selecting dairy bulls. When Rendel arrived in Australia, plans to develop tropically adapted dairy cattle were already under way through the efforts of R. B. Kelley, and Rendel grasped the opportunity to apply the principles developed by him and Alan Robertson to design the first animal improvement programme in Australia based on modern quantitative genetics theory. While the AMZ programme, as it later became known, has not been a commercial success, the principles of well designed progeny testing programmes had a very considerable influence on early dairy improvement programmes in Australia. In parallel, Rendel did much to set the direction of another animal breeding programme for the tropics, based at Rockhampton and directed at the beef industry.

However, perhaps the greatest influences on Rendel while in Edinburgh were the ideas of Waddington. It was here that Rendel developed a deep commitment to the importance of developmental biology and in particular Waddington's ideas on canalisation and genetic assimilation that led to Rendel's work on the canalisation of bristles in Drosophila. Finally, Rendel formed a strong relationship with Alex Fraser. Fraser, a New Zealander and a friend of Otto Frankel, had arrived unannounced in Edinburgh to do a PhD with Waddington. Rendel encouraged Fraser to join him in Sydney when Fraser had completed his PhD. Fraser arrived slightly before Rendel and began to set up the animal genetics unit in the Division of Animal Health and Production, in preparation for Rendel's arrival. Fraser's work on wool biology, and later in Drosophila, influenced Rendel to consider interactions among the components or determinants of any production trait, such as fleece weight. One of Rendel's later important publications, with Ted Nay, reflects those influences.

Major achievements

Much insight into Rendel's scientific outlook and approach, as well as knowledge of his results, can be obtained from Canalisation and Gene Development, which is based on a series of lectures and is in consequence clear and direct. Rendel stakes his claim in the first sentence of the preface: 'This book treats development as though it were a process initiated by a major gene and regulated through the major gene's action.' (p. 9). The experimental approach involved was not novel, in that many others (e.g. Grüneberg in mice) had studied the disruptive effects of abnormal alleles of major genes in order to understand development, following the great William Harvey's advice that

Nature is nowhere accustomed more openly to display her secret mysteries than in cases where she shows traces of her workings apart from the beaten path; nor is there any better way to advance the proper practice of medicine than to give our minds to the discovery of the usual law of Nature by careful investigation of rarer forms of disease. For it has been found in almost all things, that what they contain of useful or applicable nature is hardly perceived unless we are deprived of them, or they become deranged in some way.

We have included this lengthy quotation from Harvey (1657; 1847) because it expresses so much of Rendel's own attitude. Rendel's work was informed by the philosophy set out by Waddington (1957) but instilled into his colleagues in Edinburgh over many years.

At the time he began his work, Rendel could not hope to isolate and describe the genes involved in the control mechanisms of development, so he placed his emphasis on describing and measuring how a particular major gene influenced development of a particular set of traits. He sought to describe aspects of the process as a functional relationship between some underlying gene outcome or product, M, and the trait or phenotype, P, that is, P = f(M). He called this underlying 'complex of influences taking part in the making of a phenotype', Make (M). It is not a term that has taken on, in contrast to Waddington's canalisation, the regulation of development that produces a normal outcome in the face of environmental shocks, which is widely recognised today as important with well developed theory and experimental support, including Rendel's own pioneering work (see e.g. Gibson and Wagner 2000 and Kitami and Nadeau 2002)

It is possible that Rendel was unfortunate in the timing of his fundamental as opposed to his applied work, since it is not based on attack on the problem of the nature of the gene – that is, on how DNA constitutes the genetical message, which was the fashion throughout his civilian working life – whereas developmental genetics came into its own only during his final decade. He was certainly unfortunate in trying to quantify development, through investigation of f(M), in an era when numeracy was not demanded of developmental biologists. Many found probits (Fisher and Yates 1963) rebarbative.

Rendel investigated a sex-linked gene, scute (sc), that influenced the formation of bristles on the scutellum of Drosophila melanogaster. In the wild type, there are normally four bristles in this location, and the variance about this number is very low; in Waddington's term, scutellar bristle number is canalised at four. The sc gene both reduces the number of bristles formed and increases the variance of this number. Selection for increased bristle number in animals carrying one or more sc genes produced an increased number of bristles in wild-type flies as well as those carrying one or more sc.

On the assumption that underlying variation in M is Gaussian in distribution, the probit transformation allows estimation of intervals of M corresponding to frequency classes of different numbers of bristles (0,1,2,3,4,5,...). In this way, Rendel could show that a large range of M yielded four bristles in wild-type flies, thereby quantifying the canalisation, whereas outside this range, small changes in M could produce large changes in P. Thus, the shape of f(M) could be and was determined experimentally.

A second important finding, which confirmed Fisher's work on the evolution of dominance (though Rendel disagreed with Fisher in his interpretation of Fisher's experiments; cf. Fisher 1999), was that selection of modifiers of bristle number was possible through the disruptive effect of the major gene, but the selected modifiers were not regulated by the same system, or they would not have been selectable. Rendel concluded that dominance, though a primary concept by virtue of its recognition by Mendel a century previously, was not primary in a biological sense: developmental stability was the outcome of natural selection, and dominance of the wild type a threshold effect that was a consequence of canalisation.

Rendel, having explained evolved dominance and other phenomena in this fashion, went on to consider the action of selection more generally. His discussion of the different types and consequences of selection for an intermediate phenotype is a model of clarity. One at least of his conclusions remains important: regulation is an outcome of natural selection on account of its benefits to developmental processes, not because it yields intermediate optima. From his own experimental work and a fair-minded evaluation of that of others, he showed that variability about an intermediate value could be successfully reduced by selection if that selection was carried out among animals whose variability was enhanced by a major gene such as sc, but not otherwise. He noted in contrast that many traits were not canalised, in that directional selection, whether for an increase or a decrease, was always successful. In consequence, the variability that is important in plant and animal breeding should always be assessed for canalisation before selection was undertaken.

Rendel attempted to relate his quantitative schema to gene regulation as it was understood at the time, in particular to Jacob-Monod operon theory, but the two approaches were too far apart to yield valuable results. He was closely in touch with the development of metabolic control theory by Henrik Kacser and others, and pointed out that the case of sc could not be fitted into Kacser's framework. Recently, Wagner and colleagues (Bagheri-Chaichian and Wagner 2002, Bagheri-Chaichian, Hermisson, Vaisnys and Wagner 2002, Wagner, Booth and Bagheri-Chaichian 1997) and Omholt, Plahte, Øyehaug and Xiang (2000) have brought this discussion up to date and have shown that homoeostasis and hence canalisation are not inevitable. That is, phenomena like canalisation and dominance may be the outcome of evolution by natural selection. In the case of dominance, it may arise as an ancillary outcome of direct selection on traits controlled by genes that are likely to influence dominance. In the case of canalisation, there is an interaction between direct stabilizing selection on a trait and selection for canalisation of that trait, such that if the genetic variance in the trait is reduced to a very low level, canalising selection will be ineffective. Rendel's contribution, which pointed qualitatively towards many of these conclusions, has been largely absorbed in time.

The final stage of Rendel's work was its practical application, the Eggatron being the most important example.[2]

This was a daylight-excluding poultry layer-house with automatic recording of the time of lay for every hen and the capability to vary day length as experienced by the hens. In nature, hens lay eggs daily for a number of days, yielding a clutch, and then set this clutch to hatch. Before Rendel's work, increased egg numbers had come first from the breaking of the link between laying and setting, then from reduction of the inter-clutch interval and then from increase in the number of consecutive clutches, though these stages were not necessarily recognized as separate or separately selected. There may also have been selection for clutch size, but this is often highly canalised (see e.g. Mayo 1983, Chapter 7), and we know little of clutch size in ancestral poultry. In the Eggatron, day length less than 24 hours exposed additional genetical variation in rate of lay, so that the interval between eggs could be reduced from 24 hours. Increased rate of lay was obtained by Rendel's colleagues, working to his plans. At one time, 40% of Australian layer germplasm came from the Eggatron, through dedicated application of Rendel's ideas by Bruce Sheldon and others. Sadly, commercial breeders later chose to import germplasm rather than continue advanced work based on CSIRO research, condemning the Australian industry to import replacement and export uncompetitiveness, but Rendel's approach was commercially successful.

Helen Newton Turner's work on the Booroola gene for increased fecundity in sheep was a parallel development that was as scientifically sound and yielded, through its application to meat sheep by L. R. A. (Laurie) Piper and B. M. (Bernie) Bindon, potential for increased lambing, but it has not had as yet the commercial success of the Eggatron.

Leadership

We have already mentioned some of the people who influence Rendel in his research directions, in particular Haldane and Waddington. Haldane and his family upbringing made Rendel respect intellect deeply but be no respecter of persons – authorities in particular. Though autocratic in some ways, he yet insisted that scientists must have intellectual freedom and adequate resources to pursue their ideas.

As we have already noted, Waddington had enormous, indeed unlimited, regard for his own abilities and was delighted to appoint staff who were, as it turned out, better scientists than he, such as Alan Robertson. A man of very broad interests to whom art and travel and the company of congenial intellectuals were hugely important, he never had time to supervise his staff closely, and by appointing those whom he knew and trusted from his wartime work, as well as brilliant young people in related fields, he created a scientific powerhouse in the Institute of Animal Genetics. From him, Rendel learned to appoint the most talented people he could find, to give them very general direction, and to let them work towards success or failure. The results, as might have been predicted, were mixed.

Rendel was appointed to head the new Division of Animal Genetics in a golden hour for genetics, for CSIRO, and for the industries that the Division was to serve. His success or otherwise was therefore dependent on his overall vision, and his ability to choose the people who would bring it to fruition.

CSIRO had been established in 1949 from the already successful CSIR and was widely respected for its achievements in the national interest. Indeed, in the public mind it was almost synonymous with science, and its activities were the major part of Australia's non-defence research from 1926 on (see Mayo 2002 for discussion and references). Its founding Chairman, Ian Clunies Ross, had access to everyone from the Prime Minister down, and though he died the year before Rendel took up his post as a Divisional Chief, CSIRO's standing, and consequently its funding, were unquestioned for many years.

Rendel's Division was strongly supported by the wool and beef industries, though not as strongly as was the Animal Physiology Division by wool. The history of the wool industry in the last half-century makes gloomy reading, and many have argued that the failure of breeders to take up methods proven successful in the pig and poultry industries was in part a failure of the applied-science leadership to take the work out to industry wholeheartedly and coherently. Rendel never saw technology transfer as one of his Division's primary responsibilities, so that more effort was put into this activity in tropical dairy breeding because otherwise no progress could have been achieved, than in wool breeding. State Departments of Agriculture and Primary Industry unquestionably had technology transfer, or extension as it was usually called, as part of their mandate, and Rendel approved of and supported strong collaboration between his Division and these agencies, but he never led the effort himself. He saw his role as scientific leadership, and in this role he was fearless.

While animal production research was to some extent established in the Division of Animal Health and Production by the time that Jim Rendel arrived, he had built upon it steadily in the 1950s. Arthur Dunlop had come to the Division from a PhD at Iowa State University to carry out wool research, and Helen Newton Turner was appointed to lead sheep genetics research in the Division. H. G. Turner was recruited to head a small beef research group in Rockhampton to study the relationship between production and adaptation to tropical environments. A property, Belmont, was purchased with support from the Australian Meat and Livestock Corporation to provide experimental material for this research. Later, a poultry research group, based initially at Werribee, was moved to Sydney in new facilities at North Ryde. At Badgery's Creek, south-west of Sydney, two zebu milking breeds, the Sindhi and Sahiwal imported from Pakistan, were crossed to Jerseys under the supervision of Bob Hayman. Geoff Grigg and Hymie Hoffman were moved to Sydney from Adelaide to establish molecular and developmental biology research. Alex Reisner, who had worked on Paramecium, joined the Division, as did Peter Claringbold, who subsequently rose to the position of Chief of Computing Research.

In 1959 the Division of Animal Health & Production was split into three Divisions, Animal Health, Animal Genetics (under Rendel), and Animal Physiology at Prospect (a site initially set up by Harold Carter as a wool research laboratory). By the late 1960s, Rendel had recruited additional molecular biologists, notably Hiro Sibatani and Stephen Fazekas de St Groth, as he had developed a firm conviction that the future of genetics research lay in molecular and developmental biology.

Rendel's leadership style was based on the principles of the Duke of Wellington – who believed in appointing the best people that he could find, and then not interfering with their decisions – as put into practice by Waddington. Martin (1997, Chapter 2) gives a view of one case where this did not work, but within the Division it was generally highly successful. As CSIRO grew, its procedures became more formal (some would say bureaucratic), especially in the early seventies, and Rendel had difficulties with the senior management of CSIRO because he refused to tolerate interference or calls to justify his decisions. The Division was disbanded in 1975, with the more traditional areas merged with the Division of Animal Physiology and the molecular and genetics research spun off as a separate unit, later the Division of Molecular Biology. This was unfortunate, and was based at the time on a report by Eric Underwood, who saw a need for geneticists and physiologists to spend more time talking to one another, but who saw no relevance of molecular biology to animal production research. Rendel remained with the newly formed Division of Animal Production, returning to the bench until his retirement in 1980. His outstanding scientific leadership capabilities, however, were no longer used by CSIRO.

Initially, Rendel's laboratories were centred on the University of Sydney – a model of collocation that still works very well – with field stations both nearby at Badgery's Creek, at that time only an hour's drive west of Sydney, and far away at Gilruth Plains near Cunnamulla in south-western Queensland. The mandate was the national benefit. In that great era when Australia recognised the need to expand its research effort, resources were made available to bring the basic science and its applications to sheep and poultry together, and new premises were opened on the CSIRO North Ryde campus in 1963. The poultry genetics unit was moved to North Ryde from Werribee in 1965-66. The new Division now included branch laboratories and field stations at Rockhampton (tropical beef cattle), Armidale (sheep), and Wollongbar (tropical dairy cattle) as well as those already mentioned.

The appointments Rendel made were diverse and remarkable. Some have already been discussed. We do not discuss them all, and we exclude ourselves.

Two of the earliest appointments were Alex Fraser and Bill Sobey, both colleagues in Edinburgh. Alex Fraser (elected FAA but later resigned on relocating permanently to the USA) conducted many valuable fundamental experiments on pattern formation in such cases as bristle number in mice. He contributed powerfully to the discussion of any and all topics at seminar and in the tea room. His monograph with Short on the biology of the fleece was an important piece of work, but it was not taken up by the 'practical' geneticists led by Helen Newton Turner. The failure of most of the developmental biologists to engage directly in Merino fleece genetics and breeding was a background reason for the divergence in approach in the Division between those who wanted to apply basic quantitative genetics to fleece weight and fibre diameter to 'fine the national clip' and those who wanted a subtler but inevitably slower approach that required the interactions among the components of fleece weight to be experimentally and theoretically elucidated before fining the clip.

Bill Sobey, a personal friend and colleague, carried out significant work on rabbit fleas as vectors for myxomatosis. In 1975, following the merger of Animal Genetics with Animal Physiology, he transferred to Wildlife Research to continue important work that was no longer in the production sphere.

Peter Claringbold was a veterinarian turned computer scientist, not a particularly unusual transition at a time when no computer scientists per se were being trained. Before becoming Chief of CSIRO's newly established Division of Computing Research, he was one of many who contributed to the pioneering computerization of both breeding programmes and general record-keeping for the poultry and dairy breeding programmes. He assisted Alex Fraser enormously in his early work on computer simulation of genetic systems.

H. G. Turner, never to be confused with Helen Newton Turner, led the cattle- breeding work at Rockhampton after Rendel. He was a very clear thinker, did much important work on the nexus between adaptation and productivity in the tropics, and led the development of an internationally recognised tropical cattle research centre. Turner was ably followed by John Vercoe. Both are examples of capable scientists who did not consider that good work had to be done in a metropolitan centre; without their kind, the cattle industry in northern Australia could not have become the success it is today, nor could there have been continuity in the more basic work on mechanisms of heat tolerance, tick resistance and other important traits. The Australian Academy of Technological Sciences and Engineering has identified CSIRO's tropical beef breeding as one of the major successes in agricultural technology in Australia's first 200 years of European settlement (AATSE 1988).

Emeric Binet, part of the Hungarian diaspora that has so much enriched Australia's social, intellectual and business life since the Second World War, was a mathematician who had turned his hand to genetics. Unfortunately, brain damage suffered in a motor accident made it hard for him to concentrate on work, though the force of his intellect was undiminished. Characteristically, Rendel did not seek invalidity for Binet, believing that he could still contribute through discussion. This happened, but Binet's major contribution for many was as a source of legends, such as those to do with the consequences of damage to his thermoregulatory centre. Emeric was but one of many mathematical geneticists appointed to the Division. Pre-eminent, of course, was Helen Newton Turner, whose role in the development of sheep-breeding research was pivotal and is well known in the industry. Helen very much ran her own group within Animal Genetics, ably assisted by many in her group. The most important of her colleagues, perhaps, were Arthur Dunlop and Sid Young.

Complementing the growing strength of the Division in developing animal-breeding methodologies, Rendel's efforts to strengthen molecular and developmental biology, both in his Division and in Australia, resulted in the appointments of Grigg, Hoffman and Reisner, each of whom worked in various aspects of molecular and cellular biology. Additional appointments were made in the mid-sixties.

Hiro Sibatani joined Animal Genetics as a molecular biologist. Japanese, he was an extraordinarily talented yet slightly eccentric figure whom most of his colleagues loved. He was even more interested in the Philosophy of Science than Binet and his philosophical contributions to laboratory and tea-time discussion were more profound and productive than Binet's.

Stephen Fazekas de St Groth (FAA) was a Hungarian molecular biologist recruited specifically by Rendel to bolster molecular research in the mid-sixties. He was outstandingly able and conducted important work on the influenza virus, but did not have a major impact on animal production science. He helped establish a number of techniques in the Division, such as the development and application of monoclonal antibody technology.

Many others have made important contributions to Australian science. Among these are Bruce Sheldon and Brian Yoo, who led the poultry genetics programme, Bernie Bindon and Laurie Piper, who worked together on the genetics of fecundity in sheep, and later came to lead Co-operative Research Centres in beef cattle and sheep research, and Judith Koch, a molecular biologist working closely with Stephen Fazekas.

As we have already indicated, the atmosphere at North Ryde was intellectually exciting. Morning tea could turn into a seminar, seminars could turn into non- violent pitched battles of argument, everyone was open to challenge about his or her work at any time, except perhaps during the chess games that were a permanent feature of the tea room. Rendel himself was a keen chess and (contract) bridge player, or indeed participant in anything that sharpened one's wits.

Later years

Rendel retired from CSIRO in 1980, as already mentioned. At that time retirement at 65 was mandatory, regardless of the officer's capabilities and inclinations. Rendel's intellectual vigour and enthusiasm were as great as ever, and he was in demand as a consultant and adviser on cattle breeding, as his publication list shows.

He had always been a wide reader: his children remember him reciting 'The love song of J. Alfred Prufrock' from memory, and his interests ranged from Wellington (his wife Tresham gave him a complete set of Wellington's despatches, which he greatly enjoyed), through the King James Bible to Gibbon, Wells, Virginia Woolf and Russell. He was particularly interested in the grand sweep of history of men and ideas.

He was fascinated by the problems of consciousness: its nature and its relationship to the material world. He was hostile to the idea that new properties emerged as the complexity of organization increased in the physical world ('emergentism') and was therefore led to the belief that consciousness, albeit in a primitive form, must be as fundamental to the universe as gravity or mass. This view is philosophically unfashionable but in the view of Rendel's son-in-law Sandy Robertson has interesting parallels in the work of the Australian philosopher David Chalmers (e.g. Chalmers 1996).

His daughter Jane gives us the following picture of her father: 'Creative scientific enquiry and rationality were paramount in Jim's approach to life, together with stoicism. He cast a cold eye on anything that seemed motivated by greed, cruelty, self-delusion or sentimentality and had long trained himself to keep emotion under control. As a young man on holiday in pre-war Munich he had seen that city's mayor bundled into a car never to return and had once been surrounded by a huge crowd stirred to frenzy by Adolf Hitler: "Appalling. I could only stand there and wait for them to disperse," he said later. He endured conflicts and tragedies in his personal life by immersing himself even deeper in his work. A natural shyness as well as this guardedness could at times make him seem forbidding but he was deeply attached to his family and friends.' (Jane Robertson, pers. comm. to OM 2004).

As might be expected from this description, when he came in retirement to write a book, it was on common sense in science. In the book, he set out his views on topics from relativity (where he accepted that it was an explanation at the physical level but refuted its biological implications rather as Dr Johnson dealt with Bishop Berkeley; and Rendel always heeded Johnson's sage advice: "Clear your mind of cant") to the future of humanity, from mathematical physics to consciousness. Two of the authors of this memoir read chapters as they were written and had many deeply stimulating discussions over points of disagreement. Characteristically, Rendel would modify what he had written only if he had been convinced that he was wrong; if there was doubt, he left his words to stand. It is a matter of deep regret that he had not found a publisher at the time of his death, for the book embodied profound thought, vast experience and a generous scepticism that are rarely combined in one person. Although not all the references are clear without excessive explication, a letter about the book to one of us gives a little of the flavour of his thinking:

12th July 1996  

75 Falls Road
Wentworth Falls
NSW 2782

Dear Oliver-Bombardment will depend on shape; one does not know how 'pull' will depend on shape. Perhaps the difference has already been fed in to the argument but I can't find an account of the argument.

Metaphysics and semantics get tangled together from time to time. I take Newton to mean that there is some REASON for things coming together, for an arrow leaving a bow etc.: he calls it a force and then relates forces. We can take it for granted that things do come together because they are observed to do so. Why? Not because space is curved.

Particles moving at right angles when struck, yes, thanks, I should have been less precise. With a component at a perpendicular to the path of the photon or some such phrase. Then what happens to the photon.

Common sense. I can see that common sense is not a substitute for the scientific method. The scientific method, call it what you will, is something designed to find out how things fit together; it is not the best way of explaining to the general public what scientists suppose it has discovered-for the time being. Surely as an explanation common sense is OK?

I think Heidegger has got his history wrong. Sometimes he is right but not always. There is a distinction between finding out how to do something by trial & error and then asking why does it go like that and being led by an experiment to see if a technique will work. I can't draw a line between science & technology vis a vis cause and effect.

Many thanks for your comments. I don't follow the one about plants & turbines obeying the same rules according to Newton but not according to me. If true then g is not a push.

Yrs sincerely,
Jim.'

He also produced a short proof, almost short enough to fit in a margin, of Fermat's Last Theorem. This was rather an embarrassment to him because he was well aware that many cranks had produced short proofs containing well- or ill-concealed fallacious reasoning. Two of us could not find defects, but when we showed it to a professional pure mathematician, he identified a line that he considered assumed what had to be proved. Rendel was not happy with this statement, but had still not sent the paper to a journal at the time of his death. There the matter rests, yet one feels that, since the Theorem has been proven correct, a short proof may be waiting to be discovered.

About this memoir

This memoir was originally published in Historical Records of Australian Science, vol.15, no.2, 2004. It was written by:

  • Ian Franklin, Belair, SA.
  • Geoff Grigg, Lane Cove, NSW.
  • Oliver Mayo, CSIRO Livestock Industries, SA.

Acknowledgments

We thank Sandy Rendel, Tresham Rendel, Jane Robertson, Sandy Robertson, Julia Szulerowski and Polly Woods for assistance with information on JMR's family background, life with JMR and his move to Australia. We thank George Fraser, Bill Sobey, Peter Steele and other colleagues for other assistance in the preparation of this paper.

References

  • Australian Academy of Technological Sciences and Engineering 1988 Technology in Australia 1788-1988. Melbourne, Australian Academy of Technological Sciences and Engineering.
  • Bagheri-Chaichian, H. and Wagner, G. P. 2002 Evolution of dominance through incidental selection. SFI working papers. 2002: #02-11-064
  • Bagheri-Chaichian, H., Hermisson, V., Vaisnys, J. R. and Wagner, G. P. 2002 Epistasis and the kinetics of phenotypic robustness in metabolic pathways SFI working papers. 2002: #02-09-050
  • Bennett, J. H. and Binet, F. E. 1956 Association between Mendelian factors with mixed selfing and random mating. Heredity 10 51-55.
  • Chalmers, D. J. 1996 The Conscious Mind: In Search of a Fundamental Theory. Oxford University Press.
  • Clark, R. W. 1968 J.B.S: The Life and Work of J. B. S. Haldane. London, Hodder & Stoughton.
  • Fisher, R. A., and Yates, F. 1963 Statistical Tables for Biological, Agricultural and Medical Research. 6th edn. Edinburgh, Oliver and Boyd.
  • Fisher, R. A. 1999 The Genetical Theory of Natural Selection: A Complete Variorum Edition. Edited with an introduction and notes by Henry Bennett. Oxford University Press.
  • Gibson, G. and Wagner, G. P. 2000 Canalization in evolutionary genetics: a stabilizing theory? BioEssays 22 372-80.
  • Harvey, W. 1657 letter, written six weeks before his death, to Jan Vlakfeld of Haarlem, published in The Works of William Harvey: Translated from the Latin with a Life of the Author by R. Willis. London, Sydenham Society, 1847.
  • Kitami, T. and Nadeau, J. H. 2002 Biochemical networking contributes more to genetic buffering in human and mouse metabolic pathways than does gene duplication. Nature Genetics 32 191-4.
  • Martin, B. 1997 Suppression Stories. Wollongong, Fund for Intellectual Dissent.
  • Mayo, O. 1983 Natural Selection and Its Constraints. London, Academic Press.
  • Mayo, O. 2002 Arguments concerning government's investment in research. In Liberty and Research and Development: Science Funding in a Free Society. Edited by Tibor R. Machan. pp. 1-25. Stanford, California, Hoover Institution Press.
  • McCann, D. A. and Batterham, P. 1993 Australian genetics: a brief history Genetica 90 81-114.
  • Omholt, S. W., Plahte, E., Øyehaug, L. and Xiang, K. 2000 Gene regulatory networks generating the phenomena of additivity, dominance and epistasis. Genetics 156 969-980.
  • Rendel, J. M. 1967 Canalisation and Gene Control. London, Logos Press.
  • Robertson, A. 1977 Conrad Hal Waddington. Biographical Memoirs of Fellows of the Royal Society. 23 575-622.
  • Waddington, C. H. 1957 The strategy of the Genes. London, George Allen & Unwin Ltd.
  • Wagner, G.P., Booth, G. and Bagheri-Chaichian, H. 1997 A population genetic theory of canalisation. Evolution 51 329-347.

Publications of J. M. Rendel

  1. Rendel, J. M. 1940 Note on the inheritance of yellow bill colour in ducks. Journal of Genetics 40 439-440.
  2. Rendel, J. M. 1941 Some factors influencing the weight of table ducklings and the hatchability of ducks' eggs. Empire Journal of Experimental Agriculture 9 50-56.
  3. Rendel, J. M. 1944 Symposium 'Application of genetics to plant and animal breeding'. Genetical Society Nature 153 780.
  4. Philip, U., Rendel, J. M., Spurway, H. and Haldane, J. B. S. 1944 Genetics and karyology of Drosophila subobscura. Nature 154 260-2.
  5. Rendel, J. M. 1944 Genetics and cytology of Drosophila subobscura. II. Normal and selective matings in Drosophila subobscura. Journal of Genetics 46 287-95.
  6. Rendel, J. M. 1945 Variations in weights of hatched and unhatched duck's eggs. Biometrika 33 48-58. With appendices by J. B. S. Haldane.
  7. Rendel, J. M. and Suley, A. C. E. 1948 Genetics and cytology of Drosophila subobscura. III. Transplantation of eye-buds between Drosophila subobscura and Drosophila melanogaster. Journal of Genetics 49 38-41.
  8. Rendel, J. M. 1950 Experimental analysis of the inheritance of productivity and growth in pigs. Animal Breeding Abstracts 18 235-240.
  9. Rendel, J. M. and Robertson, A. 1950 Estimation of genetic gain in milk yield by selection in a closed herd of dairy cattle. Journal of Genetics 50 1-8.
  10. Robertson, A. and Rendel, J. M. 1950 The use of progeny testing with artificial insemination with dairy cattle. Journal of Genetics 50 21-31.
  11. Rendel, J. M. and Robertson, A. 1950 Some aspects of longevity in dairy cows. Empire Journal of Experimental Agriculture 18 69.
  12. Rendel, J. M., Robertson, A. and Alim, K. 1951 The extent of selection for milk yield in dairy cattle. Empire Journal of Experimental Agriculture 19 295-301.
  13. Rendel, J. M. 1951 Mating of Ebony vestigial and wild type Drosophila melanogaster in light and dark. Evolution 5 226-230.
  14. Rendel, J. M. 1952 White Heifer Disease in a herd of dairy Shorthorns. Journal of Genetics 51 89-94.
  15. Rendel, J. M. 1953 Heterosis. American Naturalist 87 129-138.
  16. Rendel, J. M. 1953 Conclusions from some recent research into animal breeding. Journal of the Australian Institute of Agricultural Science March, pp. 2-7.
  17. Rendel, J. M. 1954 Inheritance of birthcoat in a flock of improved Welsh mountain sheep. Australian Journal of Agricultural Research 5 297-304.
  18. Rendel, J. M. 1954 The use of regressions to improve heritability. Australian Journal of Biological Sciences 7 368-378.
  19. Robertson, A. and Rendel, J. M. 1954 The performance of heifers got by artificial insemination. Journal of Agricultural Science 44 184-207.
  20. Rendel, J. M. and Kellerman, G. M. 1955 Deoxyribonucleic acid content of marsupial nuclei. Nature 176 829.
  21. Rendel, J. M. 1955 Dwarfism in cattle. The Australian Shorthorn August.
  22. Rendel, J. M. and Sheldon, B. L. 1956 Effect of cold treatment on mutation in Drosophila melanogaster. Australian Journal of Agricultural Research 7 566-73.
  23. Rendel, J. M. 1956 Cattle breeding for the tropical North. In Beef Cattle in Australia (ed. F. O'Loghlen), pp. 77-83. Sydney, F. H. Johnston Pub. Coy.
  24. Rendel, J. M. 1957 Relationship between coincidence and crossing-over in Drosophila. Journal of Genetics 55 95-99.
  25. Rendel, J. M., Robertson, A., Asker, A. A., Khishin, S. S. and Ragab, M. T. 1957 The inheritance of milk production characteristics. Journal of Agricultural Science 48 427-432.
  26. Rendel, J. M. 1958 The effect of age on the relationship between coincidence and crossing-over in Drosophila melanogaster. Genetics 43 208-14.
  27. Rendel, J. M. 1958 Natural and artificial selection. Australian Journal of Science 22 22-27.
  28. Rendel, J. M. 1959 Optimum group size in half-sib family selection. Biometrics 15 376-81.
  29. Rendel, J. M. 1959 Canalisation of the scute phenotype of Drosophila. Evolution 13 425-39.
  30. Rendel, J. M. 1959 Variation and dominance at the scute locus in Drosophila melanogaster. Australian Journal of Biological Sciences 12 524-33.
  31. Rendel, J. M. 1960 Animal improvement. Journal of the Australian Institute of Agricultural Science 26 183.
  32. Rendel, J. M. and Sheldon, B. L. 1960 Selection for canalization of the scute phenotype in D. melanogaster. Australian Journal of Biological Sciences 13 36-47.
  33. Rendel, J. M. 1961 Evolution of dominance. In The Evolution of Living Organisms: A symposium of the Royal Society of Victoria held in Melbourne, December 1959. pp. 102-110.
  34. Rendel, J. M. 1961 Consciousness: can it be explained in terms of physics? Australian Scientist 1 149-153.
  35. Rendel, J. M. 1962 The relation between gene and phenotype. Journal of Theoretical Biology 2 296-308.
  36. Rendel, J. M. 1963 Correlation between the number of scutellar and abdominal bristles in Drosophila melanogaster. Genetics 48 391-408.
  37. Moule G. R., Norman M. J. T., Jones R. J. and Rendel J. M. 1963 Development of pastures and beef cattle for northern Australia. UNCSAT, 1963. Pap. no. E/CONF.39/C/396.
  38. Rendel, J. M., Sheldon, B. L. and Finlay, D. E. 1964 Effect of homozygosity on developmental stability. Genetics 49 471-84.
  39. Rendel, J. M. 1965 Effects of genetic change at different levels. Proc. 16 Int. Congr. Zool. vol. 6. Ideas in Modern Biology. pp. 285-95. New York, Natural History Press.
  40. Rendel, J. M. 1965 Scutellar bristles in Drosophila: a comment. Heredity 20 137-8.
  41. Rendel, J. M. 1965 Bristle patterns in scute stocks of Drosophila melanogaster. American Naturalist 99 25-32.
  42. Rendel, J. M., Sheldon, B. L. and Finlay, D. E. 1965 Canalisation of development of scutellar bristles in Drosophila by control of the scute locus. Genetics 52 1137-51.
  43. Rendel, J. M., Sheldon, B. L. and Finlay, D. E. 1966 Selection for canalisation of the scute phenotype. 2. American Naturalist 100 13-31.
  44. Rendel, J. M. 1968 Genetic control of a developmental process. In Population Biology and Evolution. (ed. R. C. Lewontin) pp. 47-66. Syracuse, NY, Syracuse University Press.
  45. Rendel, J. M. 1968 Control of developmental processes. In Evolution and Environment (ed. E. T. Drake) pp. 341-9. New Haven, Yale University Press.
  46. Rendel, J. M. 1969 Model relating gene replicas and gene repression to phenotypic expression and variability. Proceedings of the National Academy of Sciences 64 578-83.
  47. Sheldon, B. L., Rendel, J. M. and Finlay, D. E 1969 Possible example of a gene affecting allelic recombination in Drosophila melanogaster. Genetics 63 155-65.
  48. Johnston, P. G., Pennycuik, P. R. and Rendel, J. M. 1970 Selection for constancy of expression of the Tabby gene in the mouse. Australian Journal of Biological Sciences 23 1061-6.
  49. Rendel, J. M. 1971 Myxomatosis in the Australian rabbit population. Search 2 89-94.
  50. Rendel, J. M. 1972. Dairy cattle in hot climates. World Review of Animal Production 8 16-24.
  51. Rendel, J. M. 1972 Breeding cattle for the Australian North. World Review of Animal Production 8 48-56.
  52. Rendel, J. M. and Binet F. E. 1974 The effect of environment on heritability and predicted selection response: a reply. Heredity 33 106-108.
  53. Rendel J. 1974 (Moderator). Round table: adaptability of farm animals to tropical conditions. 1st world congress on genetics applied to livestock production, Madrid, Spain. Vol. 2. Madrid: Editorial Garsi, pp. 211-279.
  54. Rendel, J. M. 1975 The utilization and conservation of the world's animal genetic resources. Agriculture and Environment 2 101-19.
  55. Rendel, J. M. 1976 Is there a gene regulating the scute locus on the third chromosome of Drosophila melanogaster? Genetics 83 573.
  56. Rendel, J. M. 1977 Genetic variance and selection. Proceedings of the 3rd International Congress of the Society for Advanced Breeding Research in Asia and Oceania. Animal breeding papers pp. 20 (ii) 12.
  57. Rendel, J. M. 1977 Canalisation in quantitative genetics. Proceedings of the International Conference on Quantitative Genetics, August 16-21, 1976. Ames, Iowa State University Press, pp. 23-28.
  58. Pennycuik P. R. and Rendel J. M. 1977 Selection for constancy of score and pattern of secondary vibrissae in Ta/Ta-Ta/Y and Ta/+ mice. Australian Journal of Biological Sciences 30 303-17.
  59. Rendel, J. M. and Evans, M. K. 1978 Canalisation of the action of sc' in Drosophila melanogaster. Heredity 41 105-7.
  60. Rendel J. M. and Nay T. 1978 Selection for high and low ratio and high and low primary density in Merino sheep. Australian Journal of Agricultural Research 29 1077-86.
  61. Rendel, J. M. 1979 Canalisation and selection. In Quantitative Genetic Variation. (eds J. M. Thompson and J. M. Thoday) pp. 139-56. New York, Academic Press.
  62. Rendel, J. M. 1980 Low calving rates in Brahman cross cattle. Theoretical and Applied Genetics 58 207-210.
  63. Rendel, J. M. 1981 Cattle production in the tropics and improvement through breeding. 32nd Annual Meeting of the European Association for Animal Production, 1981; No. G2.1. 11 pp.
  64. Rendel JM 1981 Adaptation of livestock to their environment. Animal genetic resources conservation and management: Proceedings of the FAO/UNEP Technical Consultation, Rome. Food and Agriculture Organization of the United Nations. Pp. 190-200.
  65. Rendel, J. M. 1983 Creating new breeds in the wet tropics. Dairy cattle breeding in the humid tropics: Working papers presented at the F.A.O./G.A.O. Expert Consultation held in Hissar, India, February 12-17, 1979. Hissar: Haryana Agricultural University, 1983. pp. 188-198.
  66. Rendel, J. M. 1984 Decline in the number of breeds, its consequences and remedies. Genetics: new frontiers. Proceedings of the XV International Congress of Genetics. Volume IV. Applied genetics. New Delhi, Oxford IBH Publishing Co. pp. 23-33.

Notes

  1. Myxomatosis, a virus specific to rabbits, the artificial release of which for a time reduced the immense damage done by these animals to the Australian envronment and rural industries.
  2. The name Eggatron was a mild play on words, mocking the names of 'big science' equipment such as cyclotron, synchrotron and phytotron.

© 2018 Australian Academy of Science

Top