Rayleigh's appointment

The original intention of the University was that the Cavendish Professorship should be a single-term appointment, but the wording of the regulations offered the possibility of extension of the position. It stated that the post was to

Again, Sir William Thomson was an obvious choice, but by now he saw his future in Glasgow. The next obvious candidate was John William Strutt, who had succeeded to the Barony as the 3rd Lord Rayleigh on the death of his father in 1873. It was not a common occurrence for a senior member of the aristocracy and major landowner to become a professional academic, but Rayleigh had already demonstrated outstanding ability in theoretical and experimental physics. He had been senior wrangler in 1865 and first Smith'sPrize winner. He had sought to improve his understanding of experimental physics, but there were limited opportunities. Fortunately, Stokes allowed Rayleigh to observe his experimental researches in his private laboratory and so he gained some familiarity with the challenges involved. By the time his name came forward as a candidate for the Cavendish Chair, he was already known for his explanation of the colour of the sky through the process of Rayleigh scattering, and he had already written profusely on a very wide range of topics in the physical sciences, including experimental researches carried out at the family home at Terling Place in Essex.

But there was more to it than simply academic prestige. The Rayleigh estates had made most of their earnings through the sale of wheat. With the opening up of the midwest American prairies for wheat production, the price of wheat had plummeted and made the Rayleigh estates unprofitable. The agricultural depression was to last for a number of years and so Rayleigh decided to change to dairy farming. In the late 1880s, ‘Lord Rayleigh'sDairies’ supplied milk for the capital (Rayleigh, 1924). Although not decisive in encouraging Rayleigh to put his name forward as a candidate for the Cavendish Chair, the appointment would certainly help the family weather the downturn in their fortunes.

Lawrence Bragg at Manchester and the National Physical Laboratory

The Vice-Chancellor of Manchester University, Sir Henry Miers, having been Professor of Crystallography, was keen to develop the discipline. In his words,

Lawrence Bragg, although only 29, was the obvious choice to succeed Rutherford as the Langworthy Professor of Physics at the University in 1919, but it was to prove a real challenge. He had never taught physics and now was in charge of a large department which had become a leader in the dynamic field of nuclear physics under Rutherford. Despite a rocky ride in getting on top of teaching and the management of the growing laboratory during his first few years, he now had his own research laboratory and concentrated upon making the discipline of crystallography an exact quantitative science. This was to be his major and distinguished contribution during his Manchester years. As expressed by David Phillips,

His research programme got off to an excellent start with his acquisition of a new Xray spectrometer and the gift of a state-of-the-art Coolidge X-ray tube from the General Electric Company at Schenectady. From his continuing studies of crystalline structures, he developed the concept of the sizes of the common ions so that interatomic distances in atomic compounds could be determined by an additive law; these dimensions agreed quite well with the measured atomic distances (Bragg, 1921).

Thomson's agenda

Prior to his appointment as Cavendish Professor, Thomson was uncertain about his ability to sustain a career in experimental physics and had the option of concentrating on theoretical studies. Once his appointment was announced, however, he put all his energies into fostering the experimental activities of all his colleagues in the Laboratory and developing his own experimental agenda through the appointment of an expert laboratory assistant, whose salary he paid from his own pocket. Ebeneezer Everett was appointed to this post in 1888 and was to remain until 1930 when he retired due to ill health. In his touching obituary of Everett, published in Nature in 1933, Thomson wrote:

For many reasons, 1895 was a turning point in the development of the Laboratory and so it is illuminating to list the titles of Thomson's papers up until that year (Table 6.1). It is apparent that his interests were wide-ranging, but increasingly there was a strong emphasis upon the conduction of electricity in gases, which dates back to his paper of 1883, the year before he was appointed to the Cavendish Chair. Before looking in more detail into these researches, let us review the role of analogy and model-building as tools for the understanding of physical phenomena.

Analogy andmodel-building

A distinctive feature of Maxwell's approach to the application of mathematics to natural philosophy was his ability to work by analogy. As early as 1856, when he was in his mid-20s, he described his approach in an essay entitled Analogies in Nature, written for the Apostles Club at Cambridge (Maxwell, 1856a). The essence of the technique can be caught in the following passage.

While the ground-breaking discoveries in nuclear physics were taking place, the seeds of many new disciplines were being sown which were to be of central importance for the future development of the Laboratory. The great quantum revolution was in full swing and this required the assimilation of quite new theoretical tools and concepts which were to be essential for the future experimental research programme. We first address how the new theoretical physics was incorporated into the Laboratory's programme before describing the new directions taken in experimental physics.

Experimental and theoretical physics

Rutherford was notoriously unimpressed by complex mathematical theory, preferring simplicity and model-building as in his ‘billiard-ball’ model for collisions between charged particles. But he was always prepared to listen and could not ignore the success of Gamow'sprediction of barrier penetration. However sceptical he may have been, he was fully supportive of Cockcroft and Walton's experiments which relied upon the correctness of the wave mechanical model of the atomic nucleus (Section 9.6). Likewise, he was impressed by Mott and Blackett's demonstration of the need to use Bose–Einstein statistics in the scattering of helium nuclei by α-particles (Section 9.3).

There were significant and friendly interactions between the applied mathematicians and the experimental physicists in the Cavendish Laboratory, despite the fact that they belonged to different faculties. The Cambridge college system helped bring experimenters and theorists together, but against this was the almost insuperable obstacle that mathematics was taught as part of the Mathematical Tripos and physics as part of the Natural Sciences Tripos. This division placed a significant academic barrier in the formal teaching of the disciplines and to collaboration between physicists and mathematicians.

The development of quantum mechanics from Bohr's pioneering discovery of the structure of the hydrogen atom in 1913, through the old quantum theory from 1913 to 1925 and to the fully fledged theory of quantum mechanics in the period 1925 to 1930 was the product of the brilliant researches of theorists largely centred on Copenhagen, Göttingen and Munich. The schools of Bohr, Born, Sommerfeld and Hilbert produced a galaxy of brilliant young theorists, including Heisenberg, Jordan, Pauli, Kramers and many others – Pauli referred to the new quantum mechanics as Knabenphysik, young man's physics.

Restructuring the Laboratory: the immediate post-war years

With the end of hostilities, Bragg faced the challenge of converting the Laboratory from an organisation supporting the war effort to a physics department which had to be restructured in the light of a number of changing circumstances.

1. • The Austin Wing could now be returned to its original purpose, which was to relieve the serious overcrowding and lack of adequate laboratory space, a legacy of the Rutherford era.

2. • Just as at the end of the First World War, the demands of the Second World War had resulted in a backlog of undergraduate and postgraduate physics students to be trained in the physical sciences. In physics, the number of undergraduates was 50% greater than it had been before the war and the numbers in the final (Part II) year almost trebled to about 120. But the greatest increase was in the number of graduate students, which now totalled 160, of which 110 were studying for the PhD degree. The result was that, even with the addition of the new AustinWing, the Laboratory was already full to capacity by 1948 and indeed overcrowded in some areas (Bragg, 1948).

3. • The huge role which physics had played in the war, in particular in national security, meant that more physicists were needed for military and civilian research and development. The Barlow Report had recommended that there was a national need to maintain this larger number of qualified physics students. These conclusions were broadly in line with Bragg's own estimate, namely that 200 to 300 professional physicists would need to be trained each year in the UK after the war (Bragg, 1942b).

4. • As alluded to in Section 11.8, the physicists returned from the war with attitudes quite different from pre-war assumptions about how research should be carried out. The urgency of research and development in all areas of support for the war effort meant that physicists had become used to essentially unlimited financial and administrative support for their activities.

Low-temperature physics

During Mott's tenure of the Cavendish Chair, the low-temperature physics research carried out in the Mond Laboratory placed it at the forefront of the field internationally. The achievements during the Bragg era were to be reinforced by Pippard's and Shoenberg's ingenious experiments and insights into the nature of the Fermi surface, by Vinen and Hall'sdiscovery of vortex quantisation in liquid helium and by Josephson's discovery of quantum tunnelling of Cooper pairs. The vision was to carry out experiments with relatively simple apparatus but great experimental skill, very much in the Rutherford tradition.

Determination of Fermi surfaces

The Fermi surface is a theoretical three-dimensional boundary in momentum space within which the conduction electrons of a metal are contained at absolute zero. For the freeelectron model this surface is a sphere, but for a real metal the momentum states are determined by band theory, and the Fermi surface often has a remarkably complex shape, having crystal symmetry within the Brillouin zone structure of the atomic lattice, which is important in determining many of the properties of the metal (Box 15.1; see also Section 12.9.2). In the early 1950s there was no experimental determination of the shape of this surface for any metal, and band-structure calculations were not good enough to predict it reliably.

In 1954, however, Pippard showed that, in the extreme anomalous limit of the skin effect, the surface resistance provides a measure of one component of the curvature of the Fermi surface, averaged around the zone on which the electrons are moving parallel to the sample surface, and therefore ‘effective’. He concluded that it might be possible, by making sufficient observations of the surface resistance in different orientations, to determine the geometry of the Fermi surface, at least in the fairly simple case of a metal such as copper.

He carried out this programme during the academic year 1955–56, which he spent on sabbatical leave at the University of Chicago, where Morrel Cohen had arranged for the Institute for the Study of Materials to grow a very large single crystal of copper. The success of the experiment depended on cutting extremely smooth surfaces through the crystal at different angles, and this was expertly performed at the institute.

Where better to start than with the definition of physics from the Oxford English Dictionary: The Definitive Record of the English Language:

Most twenty-first-century physicists would find this definition sufficiently allencompassing to describe what physics is about.Many, including the present author, would argue that the extensive nature of the subject and its emphasis upon model-building and problem-solving enables it to embrace a vastly wider range of cognate fields, including chemistry, biology, medicine and economics. But the mid-nineteenth-century scientist, and in particular one associated with Cambridge University, would have formed a very different view.

Discoveries in physics, 1687 to 1874

It is enlightening to study a time-line of major discoveries and advances in what we would now call physics, as well as selected topics from cognate disciplines such as chemistry, mathematics and particularly technology. The items in Table 1.1 span the period from the date of publication of Isaac Newton's Philosophiæ Naturalis Principia Mathematica (Newton, 1687) to the opening of the Cavendish Laboratory in 1874. The contents of the table are necessarily subjective but provide a broad-brush picture of the standard history of physics. The country and location where the advances were made, as well as those which originated in Cambridge, are also indicated.

There are a number of striking features of Table 1.1. There is no question about the overwhelming impact of Newton's laws of motion and of gravity of 1687. Newton and his followers promoted vigorously the corpus of Newton's great works in the sciences of mechanics, statics, dynamics, gravity and optics. Nowhere was this more assiduously practiced than in his alma mater, Cambridge University, where, well into the nineteenth century, Newton's approach to the laws of motion were taught to the exclusion of the more powerful mathematical methods of analysis which had been developed in continental Europe.

Pure and mixedmathematics at Cambridge

The Galilean revolution culminated in Newton's great Philosophiæ Naturalis Principia Mathematica of 1687. The words of the title are significant: Newton had enunciated the ‘mathematical principles of natural philosophy’. In the course of his elucidation of the basic laws of mechanics, dynamics and gravity, he invented his own version of the differential calculus, the method of fluxions. In keeping with the conventions of the time, however, Newton wrote the Principia entirely in geometrical terms. Newton's book on the Method of Fluxions was completed in 1671 but only published posthumously in 1736. In the meantime, Gottfried Leibniz had independently developed his own version of the calculus about 1673, his book being published in 1684, 50 years before Newton's text. An unpleasant priority dispute developed between Newton and Liebniz. From our present perspective, the significant aspect of the different, but equivalent, approaches was that Newton's fluxions remained the preferred technique in England, particularly in Cambridge, while in continental Europe Leibniz's notation and techniques were adopted and are those commonly used today.

To understand the development of physics and mathematical physics in Cambridge, an important distinction has to be made between two different approaches to the discipline of mathematics. Francis Bacon and Jean d'Alembert made a clear distinction between pure mathematics and what was called mixed mathematics (Brown, 1991). According to d'Alembert's tree of knowledge, pure mathematics consisted of arithmetic and geometry, the former being divided into elementary arithmetic and algebra, which in turn included elementary algebra and differential algebra, the latter including the differential and integral calculus. Geometry included elementary geometry as applied in the fields of the ‘Military, Architecture and Tactics’, and transcendental geometry, which concerned the theory of curves. In contrast, mixed mathematics had as its domain

1. • mechanics, including statics and dynamics

2. • geometric astronomy, which included cosmography, chronology and gnomonics, meaning the science of sundials

3. • optics, acoustics and pneumatics

4. • the art of conjecturing, which meant the analysis of games of chance.

In modern terms, the distinction between pure and mixed mathematics corresponds more or less to that between pure and applied mathematics.

During the second half of the eighteenth century and the nineteenth century, continental mathematicians made very significant advances in the field of analysis, meaning the branches of pure mathematics which include the theories of differentiation, integration and measure, limits, infinite series and analytic functions.

These plans (Figures A.1–A.9) and the accompanying captions are taken from the series of papers prepared in 1936 by G.S. Graham-Smith of the Department of Pathology and are now housed in the Cambridge Collection in the Cambridge City Library. The originals contain information about the development of the Downing site, but these are not included in this summary, which concentrates upon the development of what became known as the New Museums site and which was to be the site of the Cavendish Laboratory until the move toWest Cambridge in 1974. All references to buildings at the New Museums site are included, illustrating the continual construction and movements of departments on the site. Each map shows the site in a particular year, the captions listing the significant changes since the previous map.

The buildings associated with the Cavendish Laboratory and the Jacksonian Professor are indicated with blue shading. Salvin's building, which occupied the central region of the site, is shown in red, and Fawcett's building in green. Graham-Smith's original lettering has been preserved. The subsequent development of the site from the point of view of the Laboratory is summarised in Figure 5.3.

The interregnum between Pippard's resignation from the Cavendish Chair in 1982 and Sam Edwards’ assumption of the position in 1984 was covered by Alan Cook as Head of Department from 1979 to 1984. On his appointment, Edwards took on the role of Head of Department for the next five years. Unlike Pippard, Edwards had been deeply involved in national and international science politics for many years. He had served as a member of the Council of the European Physical Society from 1969 to 1971. He had been a member of various committees of the Science Research Council since 1968 and of the Council'sScience Board since 1970. In 1971 he was appointed a member of the University Grants Committee and was then Chairman of the Science Research Council from 1973 to 1977. This was followed by his Chairmanship of the Defence Scientific Council from 1977 to 1980, and he was Chief Scientific Adviser to the Department of Energy from 1983 to 1988. He had also served as Vice-President of the Royal Society and of the Institute of Physics, and had been President of the Institute of Mathematics. Thus, he had a very wide range of contacts in government and industry and used that experience to begin a major expansion of the Laboratory's activities, to remarkable effect. He was famous for hosting dinners for senior figures in industry and government in his college, Gonville and Caius College, where he had accumulated a superb, and large, wine collection. When I took over as Head of the Laboratory in 1997, his only advice to me was: ‘Have dinners!’

Expansion of the Laboratory's programme

During the Pippard era, the numbers of staff members remained roughly constant (see Figure 16.5(a)). New initiatives were needed and this was brought about largely through the vision of Edwards during his five-year term as Head of Department. The funding pressures on the University with the gradual erosion of support for research and the universities meant it was a major challenge to increase significantly the numbers of tenured academic posts, despite the ‘New Blood’ scheme initiated by the government to regenerate research and teaching activity in the universities.

Nevill Mott's appointment as Cavendish Professor in 1954 marked the beginning of further changes of emphasis in the scientific direction of the Laboratory. While he worked closely with experimentalists, he carried out no experimental physics research himself, despite being Cavendish Professor of Experimental Physics. He was unquestionably the leading UK theorist in solid state physics and he was to consolidate and expand considerably both theory and experiment within the Laboratory.

Mott's pre-Cavendish days

Mott's pedigree in physics was impeccable. His parents, Charles Francis Mott and Lilian Mary Reynolds, had been research students in the Cavendish Laboratory, both appearing on the 1904 photograph of Cavendish graduate students. Mott read mathematics at St John'sCollege, Cambridge in 1924, which he completed with high distinction. His undergraduate studies spanned the dramatic years when quantum mechanics was discovered with its profound implications for physics at the atomic level –Mott was determined to master the new discipline. He was in the same college as Dirac and, despite Dirac's notorious reticence, benefitted from the presence of the only quantum theorist in Cambridge on the same level as the great European pioneers. Mott's PhD supervisor was Ralph Fowler, who arranged for him to spend the autumn of 1928 working with Bohr in Copenhagen. In the same year, he made his important contribution on the scattering probabilities of particles which obey Bose–Einstein statistics (Mott, 1928). Mott's predictions were confirmed by Chadwick'sexperiments (Blackett and Champion, 1931), which greatly impressed Rutherford (Section 9.3).

In 1929, Mott accepted Lawrence Bragg's offer of a lectureship at the University of Manchester, remarking in his autobiography that ‘Cambridge and Manchester were the only worthwhile schools of physics in the UK (with Bristol just beginning …)’ (Mott, 1986). Although he only stayed a year at Manchester, it was an important year for Mott intellectually. As he put it,

Pippard as Cavendish Professor

Unlike Bragg, Rutherford and Mott, Brain Pippard was an ‘internal’ candidate for the Cavendish Chair. He had spent most of his academic career in the Cavendish Laboratory where he had carried out outstanding research in low-temperature physics (Sections 12.9.2 and 15.1.1). As recounted in Section 13.5, he had been deeply involved in planning the move of the department toWest Cambridge and in the detailed design of the new buildings. In addition, in 1965 he became the first president of the new graduate college, Clare Hall, where again he was fully involved in the planning and construction phases of the splendid new buildings designed by Ralph Erskine and formally opened in 1969. Mott regarded Pippard as his natural successor as Cavendish Professor and this duly came about in 1971 on Mott's retirement.

During the 11 years of Pippard's tenure of the Cavendish Chair, there were numerous changes in the way the Laboratory was managed in the light of changing circumstances. With the doubling in size of the research and teaching activity under Mott, the Head of Department could no longer assume the commanding role of a Rutherford, but rather acted as the chairman of a ‘company’, the main role being to keep the whole operation on the road. In addition, Pippard was temperamentally quite different from his predecessors. He was very much a hands-on experimental physicist who prided himself on not having held a Research Council grant and asserted that he had never done any experiment which could not be built in the Laboratory workshops. These attitudes were very much in the experimental physics tradition established by J.J. Thomson and Rutherford. Pippard often expressed provocative views, as is illustrated by his banquet speech at an IBM international conference on superconductivity in 1961. John Waldram and I wrote in his biographical memoir for the Royal Society: