MacTutor
Sydney Chapman, geophysicist and natural philosopher, was born at Eccles in Lancashire on January 29, 1888, and died at Boulder, Colorado, on June 14, 1970. During his long and active life, he made outstanding advances in the three fields that mainly engaged his interests: the kinetic theory of gases, geomagnetism, and the upper atmosphere; and he exerted great influence on the scientific world at large.
Early on, Chapman's father had resolved to give his son the best education he could afford and sent him to an 'unusually scientific-grade school' where his interest in science was first kindled. A friend of his father advised that Chapman should enter a technical institute, now Salford University, and he spent two years there. A kindly Scot, and one of our first Ph.D.s, a Dr. B. Prentice, who had taken a fatherly interest in Chapman, encouraged him to sit for a scholarship to Manchester University As he relates in his reminiscences, the county of Lancashire in which he lived offered 15 university studentships a year. He was placed 15th on the list and, he says, 'I sometimes wonder what would have happened if I'd hit one place lower'. At Manchester he read engineering and graduated in 1907; but after this his interests turned to mathematics and Horace Lamb, the professor at Manchester, told Chapman that there were good prospects for an applied mathematician, but should he wish to become one he should go to Cambridge. Accordingly, Chapman sat for, and gained, an open scholarship to Trinity College, and went up in 1908. He became a Wrangler two years later and, having sat for the Mathematical Tripos in his second year, he had to stay at Cambridge for a third year to get his degree.
About this time, he began to have ambitions to do mathematical research, and he relates how he kept a look-out in reading books in the hope of finding a subject capable of further development. Encouraged by Littlewood, he chose the theory of summable series. However, as yet, he was uncertain whether he wanted to be a pure mathematician or an applied mathematician. He approached Sir Joseph Larmor, whose lectures he had attended at Cambridge, who promised to write to him after he had taken his examinations. Larmor then sent him some papers by Knudsen on the capillary flow of gases of low density, and this undoubtedly aroused his interest in the kinetic theory of gases, in which he was to predict the important phenomenon of thermal diffusion—a prediction in which he took great pride. However, the main line which
Chapman was to follow in his researches was finally settled by an unexpected opportunity. Returning one Sunday afternoon to his rooms at Trinity he found the Astronomer Royal, Dr Frank (later Sir Frank) Dyson, waiting for him. Dyson interviewed him (on Sir Joseph Larmor's recommendation) and a few days later offered him one of the two posts of Chief Assistants at Greenwich Observatory. Chapman knew little astronomy, but one of the first tasks which Dyson entrusted to him was to supervise the building of a new magnetic observatory with a view to making better observations. As Chapman says in his reminiscences, having got more accurate data he began also to consider the results. It was undoubtedly the beginning of his interest in the study of the Earth's magnetism, or Geomagnetism, as he was to rename the Science, an interest which was to last until the end of his life. He transformed the science of geomagnetism from a mere collection of articles scattered in the literature into an organic whole, which is so masterfully surveyed in the great treatise written jointly with Julius Bartels under the title of Geomagnetism. Chapman was not greatly drawn towards astronomy as such, and this may be surprising since at that time Eddington was the other chief assistant at Greenwich. His excursions into pure astronomy were few and far between, being confined almost entirely to star counts in which he collaborated with Melotte.
In 1914, Chapman returned to Cambridge as a lecturer in mathematics; he had been awarded a Smith's Prize in 1913, and in the same year he was elected to a Fellowship at Trinity College, which he held until 1919. The start of the First World War, however, depleted the student ranks at Cambridge, and because the staff at Greenwich was also reduced, Chapman was asked to return to Greenwich in 1916, on leave of absence from his college, to help out at the Observatory, taking the place of Spencer Jones. He stayed there until the end of the war.
During his first stay at Greenwich, Chapman had, in his spare time, helped run a boys' club. On his return to Greenwich in 1916, his pacifist views and his association with the local Labour Party made him unwelcome at the boys' club, and he took charge of children's work for the Labour Party instead. He was much affected by the unpopularity of his views at this time, which depressed him deeply. After the war, he returned to Cambridge, but his stay there was short; in 1919, he was asked to succeed Sir Horace Lamb, his former teacher at Manchester, as Professor of Mathematics. They overlapped during the 1919-20 session.
Chapman's stay at Manchester lasted for five years until 1924, when he was invited to succeed A.N. Whitehead as Chief Professor of Mathematics at Imperial College, London. The Department of Mathematics, he found, was quite different from what he had expected and not what the Rector of Imperial College, Sir Thomas Holland, or the Governing Body, wished it to be. Chapman had the difficult and unenviable task of having to breathe new life into what was, in effect, a poorly equipped Department of Engineering Mathematics. In his quiet, but firm, way he gradually transformed the Department, not without some considerable opposition from some established members of the Department, who rightly saw this as a threat to their chance of further advancement. In the transformation he was greatly aided by Professor H. Levy, who was a tower of strength, and by the time Chapman left the Department of Mathematics in 1946 it had acquired a considerable reputation in scientific circles. He had attracted several distinguished mathematicians, G. Temple, F. de R. Kronig, W.H. McCrea and W.G. Penney (now Lord Penney and Rector of Imperial College). But the emphasis on teaching still lay on the side of applied mathematics, and one of us [V.C.A.F.] can remember questioning him once as to why the study of pure mathematics at Imperial College was not as extensive as at other London colleges; as usual, he provided the irrefutable answer, namely, that, as its name implied, the Imperial College was a College of Science and Technology, emphasizing the word Technology.
Chapman's services were sought by the War Office during World War II. By this time, Chapman had modified his views on pacifism, considering that his earlier views had been too unrealistic and idealistic. He undertook scientific work connected with military operations, showing determination in standing up to the generals if necessary. But he worked very hard, never sparing himself, and never wasting a moment of his time. When he left in 1945, he mentioned to a friend over the telephone that he 'felt like a free man again.'
In 1946, the University of Oxford invited him to succeed A.E.H. Love as Sadleian Professor of Natural Philosophy. Here he found an atmosphere totally different from the more mundane world at Imperial College. He certainly enjoyed the gracious living there, but sorely missed the secretarial assistance, which was practically unheard of at Oxford in those days. At Oxford, as at Imperial College, he stimulated research and attracted several students, as well as some distinguished scientists, to work with him.
He retired from Oxford in 1953 to take up two posts in America, where he had been a frequent visitor during the post-war years. One was as Advisory Scientific Director at the Geophysical Institute of the University of Alaska and the other as a research worker at the High Altitude Observatory in Boulder, Colorado. Here he found great opportunities to develop a research school in geomagnetism, in which field he was now recognized as the leading authority. One of his most distinguished pupils in Alaska was S.I. Akasofu, whom he was particularly proud to have discovered. He was also fortunate to find Dr. C.T. Elvey, Director of the Geophysical Institute in Alaska. Their close cooperation led to considerable advances in auroral research.
In 1922, Chapman married Katherine Nora, daughter of A.E. Steinthal, treasurer of the University of Manchester. They had one daughter and three sons, one of whom, Robert, has since become a very distinguished architect. Chapman was very happy in his family life, in which he was much helped by a kind and understanding wife who, in later years, accompanied him on many of his travels around the world. As is well known, Chapman was a tireless and intrepid globetrotter, especially in his younger years, when he would travel almost anywhere and in any weather by cycle. Perhaps one might mention the fact that in 1939 he rode from Montreal to Washington to attend the International Union of Geodesy and Geophysics, and in 1954 he walked several miles from the center of Rome to the European Union, where the meetings were held, to give his Presidential Address to the same Union He caused some concern among the officials of the Union, for he arrived only a few minutes before he was due to speak; yet he spent these remaining minutes practicing the pronunciation of the opening sentences of his address, which had been prepared in Italian.
His tastes were simple, but he was well read and erudite in many fields and was a keen lover of the arts. Although he appeared diffident, he had a kindly disposition and a great liking for people, in whom he took great interest. In his later years, especially after his visits to America, he mellowed, and his quiet charm and directness won over many American colleagues who respected and revered him.
Chapman greatly enjoyed cooperation with other scientists; and it would seem appropriate at this point to mention Chapman's long association with the late Julius Bartels, which began in 1925, when the latter was spending a year in England. They soon became firm friends and delighted in each other's company. In 1929, Chapman was awarded the Adams Prize at the University of Cambridge for an essay on Terrestrial Magnetism, an award which carried with it the condition that the essay should be published in book form. The writing of the book proceeded very slowly until Chapman invited Bartels to join him as a collaborator. Bartels agreed and made his own very real contribution to the book, which appeared in 1940, just after the outbreak of the Second World War, under the title of Geomagnetism. Much work has been done, and many new discoveries have been made since the book was published, some vindicating Chapman's own theories. Yet to this day, the work remains as fresh as when it first appeared, and it is still the standard book on the subject.
Chapman's cooperation with E.A. Milne was brief but, nevertheless, important. They wrote a joint paper on the composition of the upper atmosphere and also on the penetration of auroral particles in the Earth's atmosphere. They were also largely instrumental in promoting the use of vector methods in mathematical physics, to which, except at Cambridge, there was considerable resistance at the time.
Chapman's association with Cowling, a pupil of E. A. Milne, who had himself been a pupil and colleague of Chapman, began in 1929, soon after Cowling had demolished a theory that Chapman had put forward to explain the supposed radial limitation of the sun's magnetic field, a phenomenon later proved to be illusory. Chapman acknowledged the validity of Cowling's criticism, and it is evidence of Chapman's magnanimity that he offered Cowling a post on the staff at Imperial College. Chapman wished to expound the Chapman-Enskog theory in book form, but the pressure of other work convinced him that he must find a collaborator. After earlier abortive attempts, Chapman asked Cowling to join him in this task, and the book finally appeared in 1939 under the title Mathematical Theory of Non-Uniform Gases.
As has already been mentioned, Chapman exerted great influence on the scientific world at large Because of his broad experience and special gifts, it was altogether appropriate that, at its inception, Chapman should have been invited to become the President of the Special Commission for the International Geophysical Year. He was largely responsible for the planning, shaping and conduct over long periods of the work of the I.G.Y. until its completion. As ever, Chapman would be the first to insist that the work was the collective efforts of many hands, but it is doubtful whether the confidence which such a broad scientific venture required could have been created without his own contribution and authority. He had always taken a leading part in the affairs of the International Union of Geodesy and Geophysics, and it would be difficult to overestimate the debt which the Union owes to him. He served as President of its filial associations for Meteorology (in 1936-48) and Geomagnetism and Aeronomy (in 1948-51), and later served as President of the Union itself from 1951 to 1954.
As was to be expected, Chapman was also greatly active and prominent in the affairs of learned societies in this country. He was a prominent member at the meetings of our Society and took great interest in all its activities; his main contribution was undoubtedly the development of its geophysical interests, which he regarded as overlapping with solar work. He must have contributed largely to the success of the Geophysical Supplement to the Monthly Notices, which was later enlarged into The Geophysical Journal. He was President of the Society from 1941 to 1943 and its Gold Medalist in 1949. He was President of the London Mathematical Society from 1929 to 1931 and was honoured by the award of the de Morgan Medal and Larmor Prize in 1944. He was also President and a Gold Medalist of the Royal Meteorological Society and of the Physical Society, which honoured him by making him the first recipient of the Chree Medal, then recently instituted by that Society. He was elected to the Fellowship of the Royal Society in 1919 at the early age of 31 and served on its Council on more than one occasion. He was awarded a Royal Medal in 1934 and the Royal Society further honoured him in 1966 by bestowing upon him their highest award, the Copley Medal. In Presidential Addresses, as well as in the named lectures which he was invited to give by these Societies, Chapman would provide an excellent review of a current topic. The lectures were not only lucid and stimulating but contained the germs of many new ideas which could be turned to advantage by young scientists with a keen eye. Indeed, it would be difficult to overestimate the debt which many a young mathematician owes to Sydney Chapman, a number of whom he launched on their careers.
It would be inappropriate here to give details of Chapman's work on the kinetic theory of gases, fundamental though it was. It concerns theoretical physics rather than astronomy or geophysics, and it has in any case been discussed elsewhere. However, mention must be made of the important applications of kinetic theory made by Chapman in both geophysics and astrophysics.
He considered the relative importance of diffusion and convective mixing in both stars (1917) and the Earth's upper atmosphere (1920 and later), showing that convection was more far-reaching in its effects than had been supposed. He was the first (1922) to apply kinetic theory to problems of viscosity, diffusion, and heat conduction in ionized gases; the formula he gave, though capable of improvement in view of a relatively crude assumption about the cut-off distance for molecular interactions, represented a remarkably good first approximation to the truth. He recognized the importance of the high electrical conductivity of stellar matter (1929). Similarly, he showed (1957) that the high thermal conductivity of the solar corona implies that its temperature and density fall off only slowly with increasing altitude, a discovery which led E.N. Parker to the explanation of the continuous outward flow of the solar wind.
Another important application was to the electrical conductivity of the ionosphere and to the effect of the geomagnetic field on it. Chapman found it difficult to reconcile the kinetic theory value of the conductivity with the value calculated by him in 1919 from the lunar semi-diurnal geomagnetic variations. Even making a generous allowance for the amplification of the lunar atmospheric tide between the ground and ionospheric heights, he found it necessary (1926, 1929) to propose a conducting layer 100–200 km wide. Later work (by Martyn & Hirono) on the theory of conductivity in a magnetic field, reducing the width required for the layer to one more consonant with that determined by rocket probes, was described by Chapman in a much later report (1956).
Chapman showed a father's pride in thermal diffusion, and ever sought its manifestations in cosmical phenomena. He showed that it is more important in ionized gases than neutral, and suggested its possible importance in explaining observed differences of composition between the solar photosphere and the corona (1958, 1959). He also hoped to explain noctilucent clouds in the Earth's atmosphere by the same mechanism, but found it inadequate; an explanation he advanced with Kendall in 1965 was that such clouds appear at the top of a high-level convective layer where water vapor convected upward condenses on meteoric dust slowly settling down through a stable layer above.
As stated earlier, it was during his period at Greenwich that Chapman became interested in the variation of the Earth's magnetic field. Balfour Stewart had suggested that the small daily variations were due to fluctuating electric currents induced in a conducting layer in the upper atmosphere by tidal motions across the Earth's magnetic field. The reason he advanced for this hypothesis was that neither the solid Earth nor the lower atmosphere was affected by the Sun in a way that could account for the changes in these variations from the sunspot maximum to the sunspot minimum. The first attempt to develop a quantitative theory of the diurnal variations was made by Schuster, who showed that the greater part of the variation was of external origin. Chapman argued that, if the theory were correct, then in so far as such variations are due to tidal action there should also be a lunar diurnal variation. In 1913, he determined the Fourier components of this variation at three stations and compared his results with deductions made from the theory, now generally known as the dynamo theory.
One of the Fourier components indicated that the electrical conductivity of the layer was higher over the sunlight hemisphere than over the dark hemisphere, a possibility which had also been recognized by Schuster. A more complete analysis by Chapman in 1919 showed that the field responsible for the solar and lunar variations could be separated into a part of external origin and a part of internal origin, approximately one-third of the whole, which he ascribed to electric currents induced within the Earth. The dynamo theory was able to explain many of the observed phenomena, but there were also some difficulties. The computed values of the electrical conductivity, after allowance was made for the supposed reduction due to the presence of the Earth's field, appeared to be insufficient to yield the currents required to explain the variations. Moreover, the semi-diurnal tidal convective motion, deduced from theory, appeared to be reversed as compared with the barometric variations at the Earth's surface.
As mentioned earlier, the first of these difficulties was eventually resolved by Martyn and Hirono almost simultaneously, using an earlier result of Cowling that a polarizing electric field could remove most of the reduction in conductivity due to the Earth's magnetic field. The second difficulty could not be resolved without detailed information about the variation of the tidal motions with height.
One of Chapman's outstanding scientific achievements was undoubtedly his work on magnetic storms. This began in 1918 with an analysis of the morphology of such storms. Following a line initiated by N.A.F. Moos at Bombay, Chapman considered the average characteristics of 40 moderate storms with sudden commencement observed at 12 observatories situated at middle and low latitudes and divided the variations into a storm-time component, , symmetrical with respect to the geomagnetic axis, and a component depending on solar time, which he denoted by , which was not necessarily in the nature of a diurnal variation He showed that the variation of , with storm-time st could be divided into two distinct phases. In the first, called the initial phase and lasting for the first few hours of the storm, he found that the effect of is to increase the horizontal field above its undisturbed mean. This increase is followed by a larger decrease lasting for about a day, called the main phase, and then a recovery to the undisturbed mean which may take several days. Chapman also showed that the form of the average magnetic disturbance did not change much within wide ranges of intensity, and that great magnetic storms passed through their two phases more rapidly than weak storms. In 1927, he fulfilled a wish expressed in 1918 to extend the analysis to polar regions and showed that the variations were characterized by a large complex disturbance, though, on the whole, there appeared to be a general decrease in the horizontal force, as in the lower latitudes.
Because the location of the currents responsible for the magnetic variations was unknown, Chapman introduced a hypothetical current system which, if flowing in a spherical current sheet concentric with the Earth, would reproduce the observed field at the Earth's surface. In 1935, he gave a more complete analysis of this current system and showed that it was especially intense in two narrow belts, one around each magnetic pole, and coinciding very nearly with the location of the auroral zone (zone of maximum auroral frequency). This strongly suggested the possibility that a part of the hypothetical current system, at least, might flow in the upper atmosphere.
In 1927, he also considered the average , part of the magnetic disturbance field, a subject to which he returned in 1952. He showed that it varied in amplitude with storm-time, that is, the time reckoned from the sudden commencement of the storm, and also to some extent with the position of the Sun relative to the station. Although his analysis of the average characteristics of a magnetic storm was to prove of considerable value for the theory of such phenomena, Chapman realized the importance of analysing individual storms, and, with Akasofu, he undertook this work not only for individual storms, but also for bays and pulsations, thus following up work that he had begun earlier with E.H.Vestine and El Wakil in the early 1930s. Chapman and Akasofu also systematically collected, in a series of papers, examples of a variety of magnetic storms which have greatly added to our knowl-edge as to how individual storms depart from the average characteristics discussed in earlier papers.
The first attempt at a theory of magnetic storms was made by Chapman in 1918 at the end of his paper on the average characteristics of magnetic storms. He attributed the source of the storm energy to the entry into the Earth's atmosphere of fast solar particles of one sign, but soon had to abandon this theory in consequence of a destructive criticism which Lindemann (later Lord Cherwell) directed at it in 1919, namely, that streams of such particles would be dispersed by mutual electrostatic repulsion long before they reached the Earth. Lindemann added to his criticism the hypothesis that magnetic storms and aurorae might be due to the interaction with the Earth's magnetic field of a stream of an ionized, but macroscopically neutral, stream of particles emitted from the Sun, but he did not attempt to solve the interaction problem in his paper.
In 1923, Chapman made a first attempt to solve this problem, and showed that the particles would move approximately together and be only slightly deflected by the Earth's magnetic field. However, Chapman's investigation, whilst correct in this respect, was faulty partly because of the limitations which he imposed on his solution at the outset by considering only the effects that arise when the stream already envelops the Earth completely, and by regarding ions and electrons as separate particles instead of a highly conducting fluid.
In 1927, one of the present writers [V.C.A.F.] became one of Chapman's first research students at Imperial College, and Chapman suggested that a fresh attempt should be made to develop the theory of magnetic storms. A re-examination of the electrical state of the solar streams during their passage from the Sun to the Earth confirmed Lindemann's conclusion that the only streams available in constructing a corpuscular theory of storms must be electrically neutral to a high degree of approximation. After several false starts, success was achieved when it was realized that an ionized gas is an excellent electrical con-ductor and that electric currents induced in the stream by its motion across the Earth's magnetic field might account for the variations during the geomagnetic storms. The stream was found to behave as if it were a perfect conductor, so that the induced currents flowed mainly on the surface of the stream. They shielded the stream from the Earth's magnetic field, and this enabled the particles in the stream to follow rectilinear paths up to the point where they entered the surface current layer. Chapman and Ferraro proved that the tubes of force from the Earth's magnetic field could not penetrate into the stream, but would exert a retarding (magnetic) pressure over its surface, the retardation being greatest over the parts of the surface nearest the Earth. Thus a cavity would be carved into the stream; this would deepen continuously until a steady state was reached in which there was equilibrium between the kinetic pressure of the stream and the magnetic pressure on the surface. This simple condition enables the location of the surface of the cavity to be calculated from knowledge of particle density and velocity, both of which are capable of measurement by space probes. The dimensions of the cavity thus calculated, of the order of a few Earth radii, agree well with the values obtained from direct measurements, made by a magnetometer borne in a space probe, of the location of the discontinuity of the magnetic field at the surface of the stream.
The compression of the Earth's tubes of force by the stream within the cavity increases the horizontal force at the Earth's surface, and this increase was identified as the increase in the horizontal force during the first phase of a magnetic storm. The main phase of the storm was ascribed to a westward ring current, the particles moving in circular orbits around the Earth. Although correct as regards the scale of the phenomenon, which placed the ring current at a few Earth radii away, the formulation of the theory was unacceptable; for such a ring current of the type considered would, in fact, be unstable. The true nature of the ring current was given by S.F. Singer in 1957, when he suggested that it consisted of charged particles trapped in the Earth's magnetic field and drifting differentially in that field, as had been first suggested by Alfvén in his interesting, but defective, theory of magnetic storms
Although Singer's exposition was incomplete, his prediction of the mode of formation of a ring current was vindicated by the discovery of the Van Allen radiation belts encircling the Earth. The theory of such ring currents was further developed by Parker and Dessler, and, with his pupil S.I. Akasofu, Chapman wrote several papers which extended the theory of the ring current in several important respects.
Chapman and Akasofu have done much to increase our knowledge of polar and auroral substorms. Contrary to some of Chapman's earlier analyses of the average characteristics of storms, they showed that, during individual magnetic storms, the variations in the horizontal force often show an asymmetry, which they ascribed to asymmetry of the ring current.
Schuster, in 1889, had already suggested that part of the variations observed during magnetic storms was due to the electric currents induced in the Earth, flowing in a uniform sphere whose radius is somewhat smaller than that of the Earth. Chapman made a first estimate of the radius and conductivity of this inner conducting sphere in 1919. His value for the conductivity differed from estimates derived from other variations, partly because this investigation did not take any account of the influence of the oceans and other conducting regions near the Earth's surface. In 1923, Chapman and Whitehead examined this matter and found that if the existing oceans were spaced uniformly over the Earth, the currents induced in them would give rise to magnetic effects comparable to those observed. In 1930, Chapman investigated the induced part of the storm-time variations of magnetic storms with A.T. Price, in whom he found a capable collaborator. In this work, later greatly extended by Price, a first inkling was obtained of the way conductivity increases with depth. Chapman also made an important study in 1935, continued in 1938 with E.H. Vestine, of the ionospheric current system associated with magnetic storms, which relied heavily on data from the first (1882-83) and the second (1932-33) polar years.
Chapman's work on lunar atmospheric tides was inspired by their possible relationship to the lunar geomagnetic variations. A small lunar semi-diurnal tide, about one-fifteenth the size of the corresponding solar tide, had already been detected in barometric records at certain tropical stations. Laplace and Airy had attempted to determine a similar tide from records at Paris and Greenwich, respectively, but without success. Chapman, following up on Airy's work, came to realize that the lack of success was due to noise produced by the solar tide and irregular barometric variations, which concealed the small lunar effect By restricting his attention to days of small barometric range, he was able, in 1918, to determine the lunar tide from 64 years of Greenwich data. The amplitude he found for the tidal variations was 0.009 mm of mercury, less than one-tenth the probable error of a single observation. The work involved, in which Chapman was greatly helped by the Greenwich staff, was enormous, and the work was successful only because of the systematic way in which he organized it.
This was followed, over the next 30 years, by some 20 papers describing determinations of the lunar tide at stations all over the Earth. In these determinations, Chapman had varying degrees of computational assistance. After a few years, hand calculating machines were replaced by Hollerith punched-card machines, but Chapman's main work was done before the advent of electronic computers. The computational method employed, as it had developed over the years, was described in a paper by Chapman and J.C.P. Miller in 1940; this was applicable whenever a small effect of known period is overlapped by larger periodic and irregular fluctuations. Because of its complexity, a simplified account, accompanied by a FORTRAN program for the calculations were made at the request of other workers in a paper by Chapman and S.R.C. Malin in 1970. The majority of our present information about the lunar barometric tide is due to Chapman and his fellow workers. In 1932, Chapman determined the lunar tidal temperature variation for Batavia, finding a roughly adiabatic variation amounting to less than 0.01°C.
Chapman, in 1924, considered the theory of solar atmospheric tides, which are partly thermal and partly gravitational in origin. Assuming that the temperature changes are due to heat convected upward, he found that the thermal and gravitational stimuli had to be about equal in effect, and that combined they could not produce a semi-diurnal tide of the size observed without resonance with a free oscillation of the atmosphere. The resonance theory, then widely accepted, is now known to be incorrect. A book by Chapman and R.S. Lindzen in 1970, entitled Atmospheric Tides, expounded both the observations and the modern theory of lunar and solar tides. It explained that, as regards the solar tides, the most important of the thermal stimuli come from heat absorbed by ozone, carbon dioxide, and water vapor at levels well above the Earth's surface and are sufficient by themselves to explain the observed tide without any close resonance
Chapman's interest in the ionosphere was closely related to his work on geomagnetic variations. He did not investigate the ionosphere directly, but sought to infer its properties from those variations. For example, he suggested in 1924 that the solar and lunar variations might originate in separate ionized layers, a suggestion which was discovered by later observations. He likewise thought that the high electrical conductivity of the ionosphere in the auroral zones might be due to direct bombardment by solar particles; however, observations during an eclipse (suggested by him) gave little support to such a belief
Chapman's most considerable contribution to the ionospheric field was in his 1931 Bakerian lecture to the Royal Society. In this, he introduced the standard Chapman layer, produced by the ionizing effect of solar ultraviolet radiation, which closely represents the lower part of the ionospheric E layer. He was able to successfully explain the diurnal and seasonal changes in this layer, and, though his ideas did not apply directly to the F layers, they offered a reasonable hope that, suitably extended, they could also explain these layers.
In the same lecture, following up on earlier work on atmospheric ozone, Chapman discussed the photochemistry of atmospheric oxygen. Despite having totally inadequate information as to atmospheric absorption coefficients, he correctly showed that oxygen must exist mainly in the atomic form above a height of about 100 km and attempted to estimate the height where the proportion of ozone in the atmosphere attains a maximum. He also made the suggestion (now generally accepted) that the green auroral line originates in three-body collisions of oxygen atoms, in which two combine to form a molecule, and the third is left excited. In a later report written jointly with W.C. Price, it was shown that atomic nitrogen is nowhere more than a minor constituent of the upper atmosphere. Chapman was not able to use later information such as absorption coefficients and atmospheric densities and temperatures; his work was a masterpiece of the art of extracting maximum information from minimum data.
Other upper-atmosphere work by Chapman was concerned with the nighttime emission of the sodium D line, and (as mentioned earlier) his work with P.C. Kendall on the formation of noctilucent clouds. He took an interest in many upper-atmosphere phenomena, and a number of the technical terms he introduced in this connection, such as scale-height, aeronomy, and mesosphere, have been generally accepted.
On June 14, 1970, Chapman suffered a heart attack followed by a cerebral hemorrhage from which he did not recover consciousness; he died two days later. His death grieved his many friends and came as a shock to those who knew how active he still was. He had become a legend during his lifetime and left behind a vast store of scientific work.
T.G. COWLING
V.C.A. FERRARO
(We have in the above drawn on our obituaries for the Royal Society [T.G.C.] and the London Mathematical Society [V.C.A.F.]. Since the list of Chapman's publications is extensive and is appended to these two obituaries, it is not repeated here.)
Early on, Chapman's father had resolved to give his son the best education he could afford and sent him to an 'unusually scientific-grade school' where his interest in science was first kindled. A friend of his father advised that Chapman should enter a technical institute, now Salford University, and he spent two years there. A kindly Scot, and one of our first Ph.D.s, a Dr. B. Prentice, who had taken a fatherly interest in Chapman, encouraged him to sit for a scholarship to Manchester University As he relates in his reminiscences, the county of Lancashire in which he lived offered 15 university studentships a year. He was placed 15th on the list and, he says, 'I sometimes wonder what would have happened if I'd hit one place lower'. At Manchester he read engineering and graduated in 1907; but after this his interests turned to mathematics and Horace Lamb, the professor at Manchester, told Chapman that there were good prospects for an applied mathematician, but should he wish to become one he should go to Cambridge. Accordingly, Chapman sat for, and gained, an open scholarship to Trinity College, and went up in 1908. He became a Wrangler two years later and, having sat for the Mathematical Tripos in his second year, he had to stay at Cambridge for a third year to get his degree.
About this time, he began to have ambitions to do mathematical research, and he relates how he kept a look-out in reading books in the hope of finding a subject capable of further development. Encouraged by Littlewood, he chose the theory of summable series. However, as yet, he was uncertain whether he wanted to be a pure mathematician or an applied mathematician. He approached Sir Joseph Larmor, whose lectures he had attended at Cambridge, who promised to write to him after he had taken his examinations. Larmor then sent him some papers by Knudsen on the capillary flow of gases of low density, and this undoubtedly aroused his interest in the kinetic theory of gases, in which he was to predict the important phenomenon of thermal diffusion—a prediction in which he took great pride. However, the main line which
Chapman was to follow in his researches was finally settled by an unexpected opportunity. Returning one Sunday afternoon to his rooms at Trinity he found the Astronomer Royal, Dr Frank (later Sir Frank) Dyson, waiting for him. Dyson interviewed him (on Sir Joseph Larmor's recommendation) and a few days later offered him one of the two posts of Chief Assistants at Greenwich Observatory. Chapman knew little astronomy, but one of the first tasks which Dyson entrusted to him was to supervise the building of a new magnetic observatory with a view to making better observations. As Chapman says in his reminiscences, having got more accurate data he began also to consider the results. It was undoubtedly the beginning of his interest in the study of the Earth's magnetism, or Geomagnetism, as he was to rename the Science, an interest which was to last until the end of his life. He transformed the science of geomagnetism from a mere collection of articles scattered in the literature into an organic whole, which is so masterfully surveyed in the great treatise written jointly with Julius Bartels under the title of Geomagnetism. Chapman was not greatly drawn towards astronomy as such, and this may be surprising since at that time Eddington was the other chief assistant at Greenwich. His excursions into pure astronomy were few and far between, being confined almost entirely to star counts in which he collaborated with Melotte.
In 1914, Chapman returned to Cambridge as a lecturer in mathematics; he had been awarded a Smith's Prize in 1913, and in the same year he was elected to a Fellowship at Trinity College, which he held until 1919. The start of the First World War, however, depleted the student ranks at Cambridge, and because the staff at Greenwich was also reduced, Chapman was asked to return to Greenwich in 1916, on leave of absence from his college, to help out at the Observatory, taking the place of Spencer Jones. He stayed there until the end of the war.
During his first stay at Greenwich, Chapman had, in his spare time, helped run a boys' club. On his return to Greenwich in 1916, his pacifist views and his association with the local Labour Party made him unwelcome at the boys' club, and he took charge of children's work for the Labour Party instead. He was much affected by the unpopularity of his views at this time, which depressed him deeply. After the war, he returned to Cambridge, but his stay there was short; in 1919, he was asked to succeed Sir Horace Lamb, his former teacher at Manchester, as Professor of Mathematics. They overlapped during the 1919-20 session.
Chapman's stay at Manchester lasted for five years until 1924, when he was invited to succeed A.N. Whitehead as Chief Professor of Mathematics at Imperial College, London. The Department of Mathematics, he found, was quite different from what he had expected and not what the Rector of Imperial College, Sir Thomas Holland, or the Governing Body, wished it to be. Chapman had the difficult and unenviable task of having to breathe new life into what was, in effect, a poorly equipped Department of Engineering Mathematics. In his quiet, but firm, way he gradually transformed the Department, not without some considerable opposition from some established members of the Department, who rightly saw this as a threat to their chance of further advancement. In the transformation he was greatly aided by Professor H. Levy, who was a tower of strength, and by the time Chapman left the Department of Mathematics in 1946 it had acquired a considerable reputation in scientific circles. He had attracted several distinguished mathematicians, G. Temple, F. de R. Kronig, W.H. McCrea and W.G. Penney (now Lord Penney and Rector of Imperial College). But the emphasis on teaching still lay on the side of applied mathematics, and one of us [V.C.A.F.] can remember questioning him once as to why the study of pure mathematics at Imperial College was not as extensive as at other London colleges; as usual, he provided the irrefutable answer, namely, that, as its name implied, the Imperial College was a College of Science and Technology, emphasizing the word Technology.
Chapman's services were sought by the War Office during World War II. By this time, Chapman had modified his views on pacifism, considering that his earlier views had been too unrealistic and idealistic. He undertook scientific work connected with military operations, showing determination in standing up to the generals if necessary. But he worked very hard, never sparing himself, and never wasting a moment of his time. When he left in 1945, he mentioned to a friend over the telephone that he 'felt like a free man again.'
In 1946, the University of Oxford invited him to succeed A.E.H. Love as Sadleian Professor of Natural Philosophy. Here he found an atmosphere totally different from the more mundane world at Imperial College. He certainly enjoyed the gracious living there, but sorely missed the secretarial assistance, which was practically unheard of at Oxford in those days. At Oxford, as at Imperial College, he stimulated research and attracted several students, as well as some distinguished scientists, to work with him.
He retired from Oxford in 1953 to take up two posts in America, where he had been a frequent visitor during the post-war years. One was as Advisory Scientific Director at the Geophysical Institute of the University of Alaska and the other as a research worker at the High Altitude Observatory in Boulder, Colorado. Here he found great opportunities to develop a research school in geomagnetism, in which field he was now recognized as the leading authority. One of his most distinguished pupils in Alaska was S.I. Akasofu, whom he was particularly proud to have discovered. He was also fortunate to find Dr. C.T. Elvey, Director of the Geophysical Institute in Alaska. Their close cooperation led to considerable advances in auroral research.
In 1922, Chapman married Katherine Nora, daughter of A.E. Steinthal, treasurer of the University of Manchester. They had one daughter and three sons, one of whom, Robert, has since become a very distinguished architect. Chapman was very happy in his family life, in which he was much helped by a kind and understanding wife who, in later years, accompanied him on many of his travels around the world. As is well known, Chapman was a tireless and intrepid globetrotter, especially in his younger years, when he would travel almost anywhere and in any weather by cycle. Perhaps one might mention the fact that in 1939 he rode from Montreal to Washington to attend the International Union of Geodesy and Geophysics, and in 1954 he walked several miles from the center of Rome to the European Union, where the meetings were held, to give his Presidential Address to the same Union He caused some concern among the officials of the Union, for he arrived only a few minutes before he was due to speak; yet he spent these remaining minutes practicing the pronunciation of the opening sentences of his address, which had been prepared in Italian.
His tastes were simple, but he was well read and erudite in many fields and was a keen lover of the arts. Although he appeared diffident, he had a kindly disposition and a great liking for people, in whom he took great interest. In his later years, especially after his visits to America, he mellowed, and his quiet charm and directness won over many American colleagues who respected and revered him.
Chapman greatly enjoyed cooperation with other scientists; and it would seem appropriate at this point to mention Chapman's long association with the late Julius Bartels, which began in 1925, when the latter was spending a year in England. They soon became firm friends and delighted in each other's company. In 1929, Chapman was awarded the Adams Prize at the University of Cambridge for an essay on Terrestrial Magnetism, an award which carried with it the condition that the essay should be published in book form. The writing of the book proceeded very slowly until Chapman invited Bartels to join him as a collaborator. Bartels agreed and made his own very real contribution to the book, which appeared in 1940, just after the outbreak of the Second World War, under the title of Geomagnetism. Much work has been done, and many new discoveries have been made since the book was published, some vindicating Chapman's own theories. Yet to this day, the work remains as fresh as when it first appeared, and it is still the standard book on the subject.
Chapman's cooperation with E.A. Milne was brief but, nevertheless, important. They wrote a joint paper on the composition of the upper atmosphere and also on the penetration of auroral particles in the Earth's atmosphere. They were also largely instrumental in promoting the use of vector methods in mathematical physics, to which, except at Cambridge, there was considerable resistance at the time.
Chapman's association with Cowling, a pupil of E. A. Milne, who had himself been a pupil and colleague of Chapman, began in 1929, soon after Cowling had demolished a theory that Chapman had put forward to explain the supposed radial limitation of the sun's magnetic field, a phenomenon later proved to be illusory. Chapman acknowledged the validity of Cowling's criticism, and it is evidence of Chapman's magnanimity that he offered Cowling a post on the staff at Imperial College. Chapman wished to expound the Chapman-Enskog theory in book form, but the pressure of other work convinced him that he must find a collaborator. After earlier abortive attempts, Chapman asked Cowling to join him in this task, and the book finally appeared in 1939 under the title Mathematical Theory of Non-Uniform Gases.
As has already been mentioned, Chapman exerted great influence on the scientific world at large Because of his broad experience and special gifts, it was altogether appropriate that, at its inception, Chapman should have been invited to become the President of the Special Commission for the International Geophysical Year. He was largely responsible for the planning, shaping and conduct over long periods of the work of the I.G.Y. until its completion. As ever, Chapman would be the first to insist that the work was the collective efforts of many hands, but it is doubtful whether the confidence which such a broad scientific venture required could have been created without his own contribution and authority. He had always taken a leading part in the affairs of the International Union of Geodesy and Geophysics, and it would be difficult to overestimate the debt which the Union owes to him. He served as President of its filial associations for Meteorology (in 1936-48) and Geomagnetism and Aeronomy (in 1948-51), and later served as President of the Union itself from 1951 to 1954.
As was to be expected, Chapman was also greatly active and prominent in the affairs of learned societies in this country. He was a prominent member at the meetings of our Society and took great interest in all its activities; his main contribution was undoubtedly the development of its geophysical interests, which he regarded as overlapping with solar work. He must have contributed largely to the success of the Geophysical Supplement to the Monthly Notices, which was later enlarged into The Geophysical Journal. He was President of the Society from 1941 to 1943 and its Gold Medalist in 1949. He was President of the London Mathematical Society from 1929 to 1931 and was honoured by the award of the de Morgan Medal and Larmor Prize in 1944. He was also President and a Gold Medalist of the Royal Meteorological Society and of the Physical Society, which honoured him by making him the first recipient of the Chree Medal, then recently instituted by that Society. He was elected to the Fellowship of the Royal Society in 1919 at the early age of 31 and served on its Council on more than one occasion. He was awarded a Royal Medal in 1934 and the Royal Society further honoured him in 1966 by bestowing upon him their highest award, the Copley Medal. In Presidential Addresses, as well as in the named lectures which he was invited to give by these Societies, Chapman would provide an excellent review of a current topic. The lectures were not only lucid and stimulating but contained the germs of many new ideas which could be turned to advantage by young scientists with a keen eye. Indeed, it would be difficult to overestimate the debt which many a young mathematician owes to Sydney Chapman, a number of whom he launched on their careers.
It would be inappropriate here to give details of Chapman's work on the kinetic theory of gases, fundamental though it was. It concerns theoretical physics rather than astronomy or geophysics, and it has in any case been discussed elsewhere. However, mention must be made of the important applications of kinetic theory made by Chapman in both geophysics and astrophysics.
He considered the relative importance of diffusion and convective mixing in both stars (1917) and the Earth's upper atmosphere (1920 and later), showing that convection was more far-reaching in its effects than had been supposed. He was the first (1922) to apply kinetic theory to problems of viscosity, diffusion, and heat conduction in ionized gases; the formula he gave, though capable of improvement in view of a relatively crude assumption about the cut-off distance for molecular interactions, represented a remarkably good first approximation to the truth. He recognized the importance of the high electrical conductivity of stellar matter (1929). Similarly, he showed (1957) that the high thermal conductivity of the solar corona implies that its temperature and density fall off only slowly with increasing altitude, a discovery which led E.N. Parker to the explanation of the continuous outward flow of the solar wind.
Another important application was to the electrical conductivity of the ionosphere and to the effect of the geomagnetic field on it. Chapman found it difficult to reconcile the kinetic theory value of the conductivity with the value calculated by him in 1919 from the lunar semi-diurnal geomagnetic variations. Even making a generous allowance for the amplification of the lunar atmospheric tide between the ground and ionospheric heights, he found it necessary (1926, 1929) to propose a conducting layer 100–200 km wide. Later work (by Martyn & Hirono) on the theory of conductivity in a magnetic field, reducing the width required for the layer to one more consonant with that determined by rocket probes, was described by Chapman in a much later report (1956).
Chapman showed a father's pride in thermal diffusion, and ever sought its manifestations in cosmical phenomena. He showed that it is more important in ionized gases than neutral, and suggested its possible importance in explaining observed differences of composition between the solar photosphere and the corona (1958, 1959). He also hoped to explain noctilucent clouds in the Earth's atmosphere by the same mechanism, but found it inadequate; an explanation he advanced with Kendall in 1965 was that such clouds appear at the top of a high-level convective layer where water vapor convected upward condenses on meteoric dust slowly settling down through a stable layer above.
As stated earlier, it was during his period at Greenwich that Chapman became interested in the variation of the Earth's magnetic field. Balfour Stewart had suggested that the small daily variations were due to fluctuating electric currents induced in a conducting layer in the upper atmosphere by tidal motions across the Earth's magnetic field. The reason he advanced for this hypothesis was that neither the solid Earth nor the lower atmosphere was affected by the Sun in a way that could account for the changes in these variations from the sunspot maximum to the sunspot minimum. The first attempt to develop a quantitative theory of the diurnal variations was made by Schuster, who showed that the greater part of the variation was of external origin. Chapman argued that, if the theory were correct, then in so far as such variations are due to tidal action there should also be a lunar diurnal variation. In 1913, he determined the Fourier components of this variation at three stations and compared his results with deductions made from the theory, now generally known as the dynamo theory.
One of the Fourier components indicated that the electrical conductivity of the layer was higher over the sunlight hemisphere than over the dark hemisphere, a possibility which had also been recognized by Schuster. A more complete analysis by Chapman in 1919 showed that the field responsible for the solar and lunar variations could be separated into a part of external origin and a part of internal origin, approximately one-third of the whole, which he ascribed to electric currents induced within the Earth. The dynamo theory was able to explain many of the observed phenomena, but there were also some difficulties. The computed values of the electrical conductivity, after allowance was made for the supposed reduction due to the presence of the Earth's field, appeared to be insufficient to yield the currents required to explain the variations. Moreover, the semi-diurnal tidal convective motion, deduced from theory, appeared to be reversed as compared with the barometric variations at the Earth's surface.
As mentioned earlier, the first of these difficulties was eventually resolved by Martyn and Hirono almost simultaneously, using an earlier result of Cowling that a polarizing electric field could remove most of the reduction in conductivity due to the Earth's magnetic field. The second difficulty could not be resolved without detailed information about the variation of the tidal motions with height.
One of Chapman's outstanding scientific achievements was undoubtedly his work on magnetic storms. This began in 1918 with an analysis of the morphology of such storms. Following a line initiated by N.A.F. Moos at Bombay, Chapman considered the average characteristics of 40 moderate storms with sudden commencement observed at 12 observatories situated at middle and low latitudes and divided the variations into a storm-time component, , symmetrical with respect to the geomagnetic axis, and a component depending on solar time, which he denoted by , which was not necessarily in the nature of a diurnal variation He showed that the variation of , with storm-time st could be divided into two distinct phases. In the first, called the initial phase and lasting for the first few hours of the storm, he found that the effect of is to increase the horizontal field above its undisturbed mean. This increase is followed by a larger decrease lasting for about a day, called the main phase, and then a recovery to the undisturbed mean which may take several days. Chapman also showed that the form of the average magnetic disturbance did not change much within wide ranges of intensity, and that great magnetic storms passed through their two phases more rapidly than weak storms. In 1927, he fulfilled a wish expressed in 1918 to extend the analysis to polar regions and showed that the variations were characterized by a large complex disturbance, though, on the whole, there appeared to be a general decrease in the horizontal force, as in the lower latitudes.
Because the location of the currents responsible for the magnetic variations was unknown, Chapman introduced a hypothetical current system which, if flowing in a spherical current sheet concentric with the Earth, would reproduce the observed field at the Earth's surface. In 1935, he gave a more complete analysis of this current system and showed that it was especially intense in two narrow belts, one around each magnetic pole, and coinciding very nearly with the location of the auroral zone (zone of maximum auroral frequency). This strongly suggested the possibility that a part of the hypothetical current system, at least, might flow in the upper atmosphere.
In 1927, he also considered the average , part of the magnetic disturbance field, a subject to which he returned in 1952. He showed that it varied in amplitude with storm-time, that is, the time reckoned from the sudden commencement of the storm, and also to some extent with the position of the Sun relative to the station. Although his analysis of the average characteristics of a magnetic storm was to prove of considerable value for the theory of such phenomena, Chapman realized the importance of analysing individual storms, and, with Akasofu, he undertook this work not only for individual storms, but also for bays and pulsations, thus following up work that he had begun earlier with E.H.Vestine and El Wakil in the early 1930s. Chapman and Akasofu also systematically collected, in a series of papers, examples of a variety of magnetic storms which have greatly added to our knowl-edge as to how individual storms depart from the average characteristics discussed in earlier papers.
The first attempt at a theory of magnetic storms was made by Chapman in 1918 at the end of his paper on the average characteristics of magnetic storms. He attributed the source of the storm energy to the entry into the Earth's atmosphere of fast solar particles of one sign, but soon had to abandon this theory in consequence of a destructive criticism which Lindemann (later Lord Cherwell) directed at it in 1919, namely, that streams of such particles would be dispersed by mutual electrostatic repulsion long before they reached the Earth. Lindemann added to his criticism the hypothesis that magnetic storms and aurorae might be due to the interaction with the Earth's magnetic field of a stream of an ionized, but macroscopically neutral, stream of particles emitted from the Sun, but he did not attempt to solve the interaction problem in his paper.
In 1923, Chapman made a first attempt to solve this problem, and showed that the particles would move approximately together and be only slightly deflected by the Earth's magnetic field. However, Chapman's investigation, whilst correct in this respect, was faulty partly because of the limitations which he imposed on his solution at the outset by considering only the effects that arise when the stream already envelops the Earth completely, and by regarding ions and electrons as separate particles instead of a highly conducting fluid.
In 1927, one of the present writers [V.C.A.F.] became one of Chapman's first research students at Imperial College, and Chapman suggested that a fresh attempt should be made to develop the theory of magnetic storms. A re-examination of the electrical state of the solar streams during their passage from the Sun to the Earth confirmed Lindemann's conclusion that the only streams available in constructing a corpuscular theory of storms must be electrically neutral to a high degree of approximation. After several false starts, success was achieved when it was realized that an ionized gas is an excellent electrical con-ductor and that electric currents induced in the stream by its motion across the Earth's magnetic field might account for the variations during the geomagnetic storms. The stream was found to behave as if it were a perfect conductor, so that the induced currents flowed mainly on the surface of the stream. They shielded the stream from the Earth's magnetic field, and this enabled the particles in the stream to follow rectilinear paths up to the point where they entered the surface current layer. Chapman and Ferraro proved that the tubes of force from the Earth's magnetic field could not penetrate into the stream, but would exert a retarding (magnetic) pressure over its surface, the retardation being greatest over the parts of the surface nearest the Earth. Thus a cavity would be carved into the stream; this would deepen continuously until a steady state was reached in which there was equilibrium between the kinetic pressure of the stream and the magnetic pressure on the surface. This simple condition enables the location of the surface of the cavity to be calculated from knowledge of particle density and velocity, both of which are capable of measurement by space probes. The dimensions of the cavity thus calculated, of the order of a few Earth radii, agree well with the values obtained from direct measurements, made by a magnetometer borne in a space probe, of the location of the discontinuity of the magnetic field at the surface of the stream.
The compression of the Earth's tubes of force by the stream within the cavity increases the horizontal force at the Earth's surface, and this increase was identified as the increase in the horizontal force during the first phase of a magnetic storm. The main phase of the storm was ascribed to a westward ring current, the particles moving in circular orbits around the Earth. Although correct as regards the scale of the phenomenon, which placed the ring current at a few Earth radii away, the formulation of the theory was unacceptable; for such a ring current of the type considered would, in fact, be unstable. The true nature of the ring current was given by S.F. Singer in 1957, when he suggested that it consisted of charged particles trapped in the Earth's magnetic field and drifting differentially in that field, as had been first suggested by Alfvén in his interesting, but defective, theory of magnetic storms
Although Singer's exposition was incomplete, his prediction of the mode of formation of a ring current was vindicated by the discovery of the Van Allen radiation belts encircling the Earth. The theory of such ring currents was further developed by Parker and Dessler, and, with his pupil S.I. Akasofu, Chapman wrote several papers which extended the theory of the ring current in several important respects.
Chapman and Akasofu have done much to increase our knowledge of polar and auroral substorms. Contrary to some of Chapman's earlier analyses of the average characteristics of storms, they showed that, during individual magnetic storms, the variations in the horizontal force often show an asymmetry, which they ascribed to asymmetry of the ring current.
Schuster, in 1889, had already suggested that part of the variations observed during magnetic storms was due to the electric currents induced in the Earth, flowing in a uniform sphere whose radius is somewhat smaller than that of the Earth. Chapman made a first estimate of the radius and conductivity of this inner conducting sphere in 1919. His value for the conductivity differed from estimates derived from other variations, partly because this investigation did not take any account of the influence of the oceans and other conducting regions near the Earth's surface. In 1923, Chapman and Whitehead examined this matter and found that if the existing oceans were spaced uniformly over the Earth, the currents induced in them would give rise to magnetic effects comparable to those observed. In 1930, Chapman investigated the induced part of the storm-time variations of magnetic storms with A.T. Price, in whom he found a capable collaborator. In this work, later greatly extended by Price, a first inkling was obtained of the way conductivity increases with depth. Chapman also made an important study in 1935, continued in 1938 with E.H. Vestine, of the ionospheric current system associated with magnetic storms, which relied heavily on data from the first (1882-83) and the second (1932-33) polar years.
Chapman's work on lunar atmospheric tides was inspired by their possible relationship to the lunar geomagnetic variations. A small lunar semi-diurnal tide, about one-fifteenth the size of the corresponding solar tide, had already been detected in barometric records at certain tropical stations. Laplace and Airy had attempted to determine a similar tide from records at Paris and Greenwich, respectively, but without success. Chapman, following up on Airy's work, came to realize that the lack of success was due to noise produced by the solar tide and irregular barometric variations, which concealed the small lunar effect By restricting his attention to days of small barometric range, he was able, in 1918, to determine the lunar tide from 64 years of Greenwich data. The amplitude he found for the tidal variations was 0.009 mm of mercury, less than one-tenth the probable error of a single observation. The work involved, in which Chapman was greatly helped by the Greenwich staff, was enormous, and the work was successful only because of the systematic way in which he organized it.
This was followed, over the next 30 years, by some 20 papers describing determinations of the lunar tide at stations all over the Earth. In these determinations, Chapman had varying degrees of computational assistance. After a few years, hand calculating machines were replaced by Hollerith punched-card machines, but Chapman's main work was done before the advent of electronic computers. The computational method employed, as it had developed over the years, was described in a paper by Chapman and J.C.P. Miller in 1940; this was applicable whenever a small effect of known period is overlapped by larger periodic and irregular fluctuations. Because of its complexity, a simplified account, accompanied by a FORTRAN program for the calculations were made at the request of other workers in a paper by Chapman and S.R.C. Malin in 1970. The majority of our present information about the lunar barometric tide is due to Chapman and his fellow workers. In 1932, Chapman determined the lunar tidal temperature variation for Batavia, finding a roughly adiabatic variation amounting to less than 0.01°C.
Chapman, in 1924, considered the theory of solar atmospheric tides, which are partly thermal and partly gravitational in origin. Assuming that the temperature changes are due to heat convected upward, he found that the thermal and gravitational stimuli had to be about equal in effect, and that combined they could not produce a semi-diurnal tide of the size observed without resonance with a free oscillation of the atmosphere. The resonance theory, then widely accepted, is now known to be incorrect. A book by Chapman and R.S. Lindzen in 1970, entitled Atmospheric Tides, expounded both the observations and the modern theory of lunar and solar tides. It explained that, as regards the solar tides, the most important of the thermal stimuli come from heat absorbed by ozone, carbon dioxide, and water vapor at levels well above the Earth's surface and are sufficient by themselves to explain the observed tide without any close resonance
Chapman's interest in the ionosphere was closely related to his work on geomagnetic variations. He did not investigate the ionosphere directly, but sought to infer its properties from those variations. For example, he suggested in 1924 that the solar and lunar variations might originate in separate ionized layers, a suggestion which was discovered by later observations. He likewise thought that the high electrical conductivity of the ionosphere in the auroral zones might be due to direct bombardment by solar particles; however, observations during an eclipse (suggested by him) gave little support to such a belief
Chapman's most considerable contribution to the ionospheric field was in his 1931 Bakerian lecture to the Royal Society. In this, he introduced the standard Chapman layer, produced by the ionizing effect of solar ultraviolet radiation, which closely represents the lower part of the ionospheric E layer. He was able to successfully explain the diurnal and seasonal changes in this layer, and, though his ideas did not apply directly to the F layers, they offered a reasonable hope that, suitably extended, they could also explain these layers.
In the same lecture, following up on earlier work on atmospheric ozone, Chapman discussed the photochemistry of atmospheric oxygen. Despite having totally inadequate information as to atmospheric absorption coefficients, he correctly showed that oxygen must exist mainly in the atomic form above a height of about 100 km and attempted to estimate the height where the proportion of ozone in the atmosphere attains a maximum. He also made the suggestion (now generally accepted) that the green auroral line originates in three-body collisions of oxygen atoms, in which two combine to form a molecule, and the third is left excited. In a later report written jointly with W.C. Price, it was shown that atomic nitrogen is nowhere more than a minor constituent of the upper atmosphere. Chapman was not able to use later information such as absorption coefficients and atmospheric densities and temperatures; his work was a masterpiece of the art of extracting maximum information from minimum data.
Other upper-atmosphere work by Chapman was concerned with the nighttime emission of the sodium D line, and (as mentioned earlier) his work with P.C. Kendall on the formation of noctilucent clouds. He took an interest in many upper-atmosphere phenomena, and a number of the technical terms he introduced in this connection, such as scale-height, aeronomy, and mesosphere, have been generally accepted.
On June 14, 1970, Chapman suffered a heart attack followed by a cerebral hemorrhage from which he did not recover consciousness; he died two days later. His death grieved his many friends and came as a shock to those who knew how active he still was. He had become a legend during his lifetime and left behind a vast store of scientific work.
T.G. COWLING
V.C.A. FERRARO
(We have in the above drawn on our obituaries for the Royal Society [T.G.C.] and the London Mathematical Society [V.C.A.F.]. Since the list of Chapman's publications is extensive and is appended to these two obituaries, it is not repeated here.)
Sydney Chapman's obituary appeared in Journal of the Royal Astronomical Society 13:3 (1972), 464-476.