by Andrew Brown
I am very puzzled about what your hot idea on phase determination is. It is worth putting everything into it if it really works, because, short of that, I rather despair of doing anything serious with the proteins.12
In 1946, Bernal arranged for Booth to go to the inaugural meeting of the American Society for X-ray and Electron Diffraction at Lake George, New York and to carry out a tour of the nascent American computer centres. He visited Harvard, the Massachusetts Institute of Technology, the Moore School in Philadelphia, and the Bell Labs in New Jersey, but the project that excited him most, although it was not as developed as some of the others, was von Neumann’s at Princeton. Bernal had persuaded Warren Weaver of the Rockefeller Foundation to contribute $400 towards Booth’s expenses for the short visit, and once Booth had returned to London, with Sage’s backing, he was able to obtain a Rockefeller Fellowship to return to the Institute of Advanced Study (IAS) at Princeton in 1947. Von Neumann’s idea for the IAS computer was to store information on a cathode-ray tube, a notion lavishly supported by the RCA record company. Booth thought that this approach would not work, and even if it did, it would be unaffordable at Birkbeck. He decided that magnetism could provide the solution to the problem of a reliable data storage device, and bought ten discs coated with an oxide, which were components of a well-known voice-recording machine. His idea was to have a rotating floppy disc, on to which a magnetic record would be imparted. Booth waited until he got back to London to try out his new invention, and through his own contacts was able to use the laboratories of the British Rubber Producers’ Association, there being no suitable equipment at Birkbeck. He later described his failure to produce what would have been the world’s first floppy disc:
I would spin the 10-inch Mail-a-Voice disc at about 3,000 RPM at which speed it would stay flat, and then move a rigidly mounted read–write head close to the surface. The theory was that the Bernoulli effect would draw the disc to a fixed distance from the head and maintain a very small air gap. Unhappily this did not occur, the attraction was perfect but the distortion of the disc surface resulted in unstable ‘flapping’ which led to eventual disintegration.13
His next attempt was much more successful: a two-inch diameter cylindrical drum coated with permanent magnetic material (nickel). The cylinder rotated under a series of read–write heads, each about the size of a matchstick, and with the associated magnetic circuitry it provided a compact and permanent way of storing digital data – the forerunner of modern hard drives. Booth gave the first demonstrations of his system in May 1948 and it was incorporated into many early computers. One of these was the All Purpose Electronic X-ray (APEX) Computer for the Birkbeck crystallography department of the early 1950s: this computer model was manufactured commercially by the International Computers and Tabulators company and outsold the combined output of all other British manufacturers (more than 120 machines).14 Much of Booth’s research was in the new field of machine translation – the use of computers to translate natural languages. This was a direct consequence of his 1947 trip to the USA, when Weaver (who was the instigator of the subject) mentioned to Booth that while the Rockefeller Foundation would not support the development of computing in London for numerical analysis, they might grant funds for non-numerical uses such as translation.15
Although Booth continued to work on mathematical problems in crystallography, his prototype computers were never of much practical use to the Birkbeck crystallographers because they did not have data storage capacity adequate for protein work. In consequence, some were inclined to take a rather jaundiced view of Booth, but Bernal, recognizing his originality and imagination, remained supportive and was instrumental in creating a new Department of Numerical Automation at the College. In 1998, Booth had the pleasure of telling an audience at Birkbeck that a precise structural determination that in the early 1940s had taken a team of two or three people three years to complete, and for which he had subsequently written a computer program, could now be run on an Intel P2 400 MHz personal computer in less than 0.1 second!
Astbury had warned Sage in 1931 not to attempt to ‘snaffle’ protein structure, but to restrict himself to the exacting analysis of the constituent amino acids and leave the overall protein molecule to him.16 In setting up his new institute, Bernal made it clear that he would not trespass into Astbury’s territory of long polymers, and he told Carlisle that they would confine themselves to small molecules of biological interest.17 In the years before the war, there had been new interest in nucleic acids, particularly DNA, which was recognized to be an active component of chromosomes. It had been established by chemists that DNA and other nucleic acids were linear polymers – long chains of nucleotides linked together. A nucleotide is a complex unit of a sugar, a base and a phosphate group. In 1938, Astbury had announced that DNA was a single-chain molecule, with the nucleotides sitting on one another like a pile of pennies.18 The nucleotides in DNA consist of a deoxyribose sugar coupled with one of four bases, plus the phosphate. Astbury and his assistant Florence Bell decided that the sugar residues and the bases (which were all ringed-structures) lay in the same plane and that like a pile of pennies or stack of plates, each nucleotide was perpendicular to the primary axis of the chain molecule.
Just as Sage had made the first determination of the shapes of individual amino acid molecules, he now thought it would be valuable to make X-ray studies of nucleotides and their component parts. The first work at Birkbeck was done by a doctoral student, Geoffrey Pitt, supervised by Carlisle. Pitt confirmed Astbury’s assumption that the pyrimidine type of base found in DNA was a flat, six-carbon-atom ring. He was the first to provide accurate data on bond lengths and angles for pyrimidine bases. A second PhD candidate, Sven Furberg, soon arrived to build on Pitt’s success. Furberg was encouraged by his chemistry professors in Oslo to spend time in Bernal’s laboratory and came to Birkbeck in 1947 on a two-year British Council scholarship. He was quiet, courteous and clever. Carlisle managed to obtain some crystals of cytidine (which is formed by the pyrimidine base, cytosine, linking to a deoxyribose sugar ring), and gave this material to Furberg to make the first X-ray study.
The cytidine crystals were prism-shaped, about 3 mm in length and 0.08 × 0.08 mm in cross-section. After X-ray exposures averaging 130 hours, Furberg managed to define the unit cell of the crystal and amassed photographs, which were exceedingly difficult to interpret. Years later, Francis Crick commented that Furberg’s was ‘a very remarkable piece of work. It was in fact such a difficult problem that if he had asked advice he would have been told that it was not possible to solve it.’19 Drawing on the established chemical data and showing great physical insight, Furberg combined two-dimensional Fourier projections into a three-dimensional structure that has stood the test of time. His main conclusions were that Pitt’s flat ring for the pyrimidine base was correct, but the five-membered sugar ring (deoxyribose) was puckered with one of its five carbon atoms lying out of the plane of the other four. He provided detailed data on bond angles and bond lengths and concluded that, far from being parallel to each other, the rings of the base and the sugar were virtually at right angles.20 Was DNA like a pile of bent pennies?
The use of rooms at the Royal Institution lasted less than a year, and in the summer of 1946 Bernal’s department was scattered to the four winds. Pitt went to Birmingham to carry out his work on pyrimidine structure, while Carlisle and his organic group made do with two rooms of a suburban house in Hendon. On one occasion, Sage arranged to drive Carlisle and his researchers from Hendon to Cambridge for a meeting. They packed into his small car and of course Bernal was bursting with ideas for what their next research might be. To illustrate them, he drew diagrams on a sheet of paper balanced on the steering wheel, as he drove.
Ehrenberg’s physics group returned to the bomb-damaged Breams Buildings. Booth did continue to work at the RI and came to Birkbeck to lecture in the evenings. That summer, a cheerful young man from Manchester, recently demobbed from the Army, answered a newsp
aper advert to become Technical Supervisor to the Crystallography Department. Tall with wavy hair, he had joined the Royal Signals Corps in 1939 and saw action in North Africa, the Middle East and Italy. His name was Stan Lenton: he would become indispensable as Bernal’s chauffeur, mechanic, and occasional butler, in addition to his paid job overseeing the laboratory staff and equipment. Bernal soon entrusted to him the role of departmental accountant. When he first arrived at Birkbeck, he asked about the whereabouts of the crystallographers, and was told, ‘that lot, you won’t see them for months’.21
Sage still drove his pre-war Austin 10, which had spent much of the war parked at various airfields, while he was overseas. Apart from gross mechanical neglect, the car’s back doors remained integral components of the vehicle only by being tied together with rope across the backseat. Lenton, who had no car of his own, offered to look after it if he could make use of it while Sage was away – a bargain that was accepted with alacrity. Even though the streets of London were not crowded in those days, Sage would sometimes arrive at work late, looking sheepish and admit to a driving mishap. It was Lenton’s job to retrieve the car. On one occasion, there had been a collision with a London taxi. When Lenton arrived at the scene, he found the cabbie studying geometrical patterns drawn in the dirt on his taxi. Sage had demonstrated, using angles of incidence and deflection and Newton’s laws of motion, how the accident had not been his fault: not only was the cabbie satisfied, he told Lenton he was never going to clean his taxi.
A second incident required quick thinking from Lenton. Sage had broken down at Piccadilly Circus and simply abandoned the car. Lenton arrived to find two cross policemen directing the traffic around the obstacle. After Lenton explained that the car was not his, but belonged to a distinguished university professor, the police were not impressed. One of them said, ‘He is going to get done anyway.’ But, said Lenton, this professor had been Lord Louis Mountbatten’s chief scientific adviser in the war and had been responsible for D-Day planning. He was sure that Mountbatten would have to be informed if the police were going to take any action. The policemen helped Lenton push the car out of the way, and watched as he got it started with a bent paper clip in the ignition – Sage having lost the key.22
In 1947 London University granted Birkbeck the use of two Georgian houses that, prior to the Blitz, were part of a terrace just north of the Senate House. Numbers 21 and 22 Torrington Square would house Birkbeck’s Crystallography Department for the next twenty years. Number 22 was propped up by large timbers; there was a gaping hole in the back wall, and the builders were in the middle of making repairs when the first scientists moved in. The apparatus moved back from the RI had to be kept under tarpaulins to protect it from brick dust and rain. Each house had four storeys plus a basement. The rooms, arranged around the staircases, were fairly small but would serve as laboratories. The top two floors of No. 21 were taken by the chemistry department, who would not prove to be ideal co-tenants. The original servants’ quarters right at the top of No. 21 became Bernal’s flat.
The provision of living accommodation for Sage came at a useful time. Margaret Gardiner was aware enough of his philandering to decide that she would no longer have him living with her. They remained fond of each other, and Margaret accepted that it was impossible for him to love only one woman: she continued to believe that ‘in his curious way, Desmond was a very faithful person’.23 For Martin, who had seen more of his father during wartime than was the norm, there came a realization that his father was unusual in not coming home every night. Sage went back to his old haunts and found himself the main attraction to a circle of women, who were as open about sharing their favours with him as they were in sharing their politics. Successive young research workers were warned that if they heard footsteps on the wooden staircase going up to the professor’s flat, they should cover their ears. At social gatherings, some of the young wives or girlfriends of his research team would sense lasciviousness in the professor’s attitude towards them.
Sir Lawrence Bragg performed the grand opening of the Biomolecular Research Laboratory, 21–22 Torrington Square, on 1st July 1948. A booklet was produced by Birkbeck College to mark the occasion, with an introduction from the director, J.D. Bernal. The four assistant directors were listed with their research teams (15 individuals). The support staff numbered nine, headed by Miss Anita Rimel, who had replaced Brenda Ryerson as Bernal’s secretary at Birkbeck during the war. Anita was a dogged communist, with advanced views on women’s rights.24 One of eleven children of a Jewish food merchant, she was stocky and had frizzy dark hair. She was devoted to Sage and eager to make herself known to his wide circle of colleagues – which she did, by adding postscripts of her own to his dictated correspondence. One name that might well have appeared as a researcher on the X-ray analysis team, but did not thanks to Anita, was Francis Crick. He had decided that he wanted to make the transition from physics to biology, following a wartime career designing acoustic and magnetic mines. After a little preliminary investigation, including reading The Search, Crick decided that Professor Bernal at Birkbeck would be the best mentor for him. He paid a visit but got no further than Miss Rimel, who struck him as ‘an amiable dragon’. ‘Do you realize’, she asked Crick, ‘that people from all over the world want to come to work with the professor. Why do you think he would take you on?’25
While Anita Rimel would become the dominant, some would say domineering, personal assistant in Bernal’s life, she was but one of a trio of women trying to keep tabs on him at the end of the war. He still needed a secretary at COHQ and there was Kathleen Watkins at the Ministry of Home Security at Princes Risborough, who would soon move to London to help Sage run the Scientific Advisory Committee at the Ministry of Works. Sir Reginald Stradling, the wartime head of the Ministry of Home Security, was now the chief scientific adviser to the Ministry of Works and appointed Bernal to the new position at the end of 1944. Sage’s brief was to take charge of research on how best to meet the urgent need for new housing in Britain. The new role would enable him to indulge his love for architecture within a state-controlled programme devoted to improving the lives of ordinary citizens. There would also be an opportunity to carry out the first systematic scientific analysis of conventional and novel building materials at Birkbeck.
In June 1941, at a time when he was preoccupied by analysing various aspects of bomb damage, Sage composed a long note on ‘Research organisation in the building industry’.26 Writing as a member of the 1940 Council (a self-appointed body to promote planning of the social environment), he called for the setting up of a Building Research Council so that the post-war need for construction could be met as well and speedily as possible. Assuming it was ‘axiomatic that structural truthfulness is as essential to great architecture as it has ever been’, he argued that:
It is a common mistake to assume that the introduction of the scientific method into the field of design tends to cramp the imaginative powers of the designer. The reverse is true, for it is the supreme characteristic of the scientific outlook that it forever looks forward.27
It was natural, therefore, that the Building Research Station (BRS) should consult Bernal about their post-war plans. While generally impressed with their forward-looking and comprehensive approach, Sage suggested that the techniques of operational research should be used to examine the utilization of existing buildings, so that better designs and improved domestic equipment could be incorporated into the new houses. Again and again he would stress the functionality of buildings: their designs should match the uses to which they would be put, but before that could be achieved, much more needed to be known about the ways people actually behaved at home, in school or at work. In his comments to the BRS director, Bernal also urged consideration of fundamental physical properties of silicates (cement, brick and glass), fibres (wood, fibre board) and metals. There was a need ‘to develop crystal structure and phase diagrams for silicates in various states’, and for the detailed study of the setting
process in cement, (including accelerating and retarding agents). Sage made an initial visit to the Building Research Station at Watford in September 1944, when he was in the middle of preparing the final reports on Overlord and getting ready for his trip to South-East Asia. Among the topics he discussed with the staff were the temperature effects, shrinkage, creep, and plasticity of cement, the use of porous materials, soil mechanics, stress in large masses, design of complex structures, and the problem of vibrations.28
Sage made promises at the Ministry of Works soon after his appointment that he would discover why cement sets and why it is subject to shrinkage and creeping. The cracking of concrete was a common and unpredictable occurrence that vexed the construction industry. Partly to educate himself, but more importantly to raise the level of interest in these rather mundane-sounding matters, Sage suggested that a meeting be held to discuss the fundamental scientific problems associated with building. The result was a one-day symposium at the Royal Institution in May 1946 on ‘Shrinkage and cracking of cementive materials’. One of the speakers described recent electron microscopy studies suggesting that cement, in its hydrated form, consisted of ‘a matrix of interlaced fibrous crystals’29 that gives concrete its cohesive properties. Water is held in the capillary-like spaces between the crystals and depending on the moisture content, the volume of the concrete is altered giving rise to internal stress and, eventually, to cracking. In ‘a short contribution to the afternoon session, Prof. Bernal commented that the fibrillar material revealed by electron microscope studies of set cement suggests an analogy with certain other fibrillar materials such as tobacco mosaic virus, and with the behaviour they show when the concentration of their solutions is changed.’30
