Fifty years ago, after the period (1911-16) during which Sydney Chapman discussed problems of the viscosity, conduction, and diffusion of gases, he gave theoretical and experimental evidence of the phenomenon of thermal diffusion, which plays an important role in the terrestrial atmosphere. In 1920, he published with Milne a long memoir on the composition, ionization, and viscosity of the atmosphere at great heights. In that paper they reviewed the subject of the nature of the outer layers of the earth's atmosphere, namely above 20 km. At that time, the portion of the atmosphere accessible to direct observations was divided in two regions, the troposphere and the stratosphere. They proposed a departure from the views then current by suggesting that mixing might extend to heights many times greater than that of the tropopause. The stratosphere, which was distinguished by the absence of any systematic temperature gradient, was described as an atmosphere in isothermal equilibrium with no convective motion. For the investigation of phenomena requiring a knowledge of the composition at great heights, the only direct evidence was given by the study of the spectra of aurorae known to occur within the range 100 km to 130 km. Molecular nitrogen was reported as the principal contributor of the auroral spectrum, together with a bright green line at 0.557 u. The source of this line was still in 1920 the subject of much speculation; it had been ascribed in 1911 by Wegener to the hypothetical gas "geocoronium" of molecular mass 0.4. In this connection one may mention the hypothesis of Vegard, which attempted to account for this line by the bombardment of crystals of solid nitrogen.
Thus, the observations, in 1920, all illustrated an important fact---that atomic and molecular oxygen, helium, and hydrogen were not known in the auroral spectrum, and even that the amount of molecular hydrogen in the atmosphere near the ground was uncertain. Furthermore, if the presence of ozone in the atmosphere was known, its amount and distribution were matters of great uncertainty.
In 1930, known facts relating to the upper atmosphere having increased rapidly in ten years, Sydney Chapman published several theoretical papers on upper-atmospheric ozone, and was preparing the 1931 Bakerian Lecture. At that time, the main known facts regarding ozone, such as its average amount (2 mm at the equator and about 3 mm in Europe), its annual variation (spring maximum and autumn minimum), and its day-to-day variation (associated with meteorological conditions) with no daily variation, suggested to Sydney Chapman the development of a quantitative theory of the equilibrium and changes of ozone in the upper atmosphere. He wrote at the time when the level of maximum ozone concentration was supposed to be at 40 to 50 km, and just after the identification of the auroral green line as due to atomic oxygen. McLennan's identification (1927) of the experimental oxygen line with the airglow line measured by Babcock and with the auroral green line proved that oxygen was present in the atomic state above 100 km during auroral displays and at an unknown height in the airglow. Chapman showed, after having considered the changing balance between the formation and destruction of ozone by ultraviolet radiation in the main layer, that above 100 km "the oxygen exists mainly in the atomic form mixed with O2, and a far smaller proportion of O3." Thus, he introduced an important aeronomic problem, namely the ratio of atomic and molecular oxygen, thirty years before the first measurements became possible by rockets.
In his Bakerian Lecture, taking into account (among other things) the existence of two ionized layers at altitudes of about 100 km and 200 km, Chapman discussed the problem of solar ultraviolet radiation and its absorption with dissociative and ionizing effects. After having considered that the green line in the airglow and aurora is emitted by neutral atomic oxygen during a transition from a metastable excited state, he expressed the opinion that "the energy is most likely to be of solar origin, and be stored up in the atmosphere during the daytime and slowly expended throughout the night." On this view the green line in the airglow must be the result of reactions between ionized or neutral particles. The main reaction accepted by Chapman in 1931 was a three-body collision when three oxygen atoms collide, two of which combine to form a molecule O2, releasing sufficient energy to excite the third to the metastable state, upper level from which the green line is produced.
During the period 1931-40, aeronomic problems were discussed every year by Chapman in a continuous development of the subject of his Bakerian Lecture. In 1931 he determined in two important papers the absorption of monochromatic radiation in an atmosphere which was either plane-stratified or on a spherical earth. This analysis led him to determine the variation and distribution of electron-density with the introduction of the recombination coefficient. The following year he studied the influence of a solar eclipse upon upper-atmospheric ionization in order to try to settle the question of the origin of the E layer by ultraviolet or corpuscular radiation.
In 1933 and 1934, in presidential addresses delivered before the Royal Meteorological Society, he introduced the problem of the atoms and molecules in the atmosphere with a new analysis of the ozone problem and of other atmospheric gases.
In 1935, he reported, with Appleton, on ionization changes during a solar eclipse, showing that the eclipse effects in the E and Fl layers occurred at the time of the optical eclipse, indicating that these layers are ionized by light.
In 1936, he published, with W. C. Price, an important account of the recent advances on the aeronomic processes in the terrestrial atmosphere. He discussed the absorption of oxygen and nitrogen and also the atmospheric processes involving oxygen atoms.
In 1937, considering that too great prominence was given to theories attributing all the phenomena of the upper atmosphere to high-speed electronic emission from the sun, Sydney Chapman established again the theory of the three-body recombination of oxygen atoms and the excitation of the green line.
In 1938, after discussing again the electric current-system to which geomagnetic storms can be attributed, Chapman extended his studies on the lunar atmospheric tide with publication of a new determination for Accra, after Honolulu, Azores, Melbourne, Canadian, and Japanese stations. After thirty years, the lunar tidal action upon the atmosphere is still engaging the attention of a Lunar Variations Committee of the Inter- national Association of Geomagnetism and Aeronomy with the Association of Meteorology and Atmospheric Physics.
In 1939, Sydney Chapman published with T. G. Cowling the famous book, The Mathematical Theory of Non-Uniform Cases. Here the fundamental formulas on diffusion, conduction, and viscosity necessary for the aeronomic studies can be found. In the same year, as a sequel to his two 1931 theoretical papers on the ionization of the upper atmosphere by monochromatic-radiation, he discussed mathematically a modification. The height-distribution of the absorption of ionizing radiation is considered if the radiation concerned is spread over a range of wavelengths corresponding to an absorption band for some constituent. Other papers dealing with aeronomic problems and published in 1939 refer to atmospheric sodium and to the chemical composition and physical constitution of the upper atmosphere.
In 1940, on the eve of World War II, Chapman and Barters published their book, Geomagnetism. Among the 1000 pages several aspects of agronomy are described.
In 1950, ten years later, we (Bates and Nucleate) were with Sydney Chapman in California, at Pasadena, where he was preparing his Presidential Address as President of the Physical Society, on the phenomena of the upper atmosphere. However, we had met him before several times at various international meetings in Europe (Belgium, England, France, Norway) where different aspects of agronomy were discussed, showing that the upper-atmosphere problems were not simple. In Pasadena, we had the privilege to discuss with him the nomenclature he was proposing in relation to atmospheric composition, temperature distribution, and ionization. There, he corrected the text of my paper on the extension of his theory of the absorption distribution of monochromatic radiation of a gas in an exponentially distributed atmosphere with constant scale-height to an atmosphere with "varying" scale-height (called graded scale-height by Chapman).
In 1951, in Brussels, Sydney Chapman was elected President of the International Union of Geodesy and Geophysics (IUGG). In 1954 at its final plenary session in Rome, he announced the adoption of the new name, the International Association of Geomagnetism and Aeronomy, for what was previously the International Association of Terrestrial Magnetism and Electricity.
In 1953, we were in close association when he became President of the Special Committee for the International Geophysical Year and I became the Secretary General. During several years (1953-60) we worked together in the organization of the International Geophysical Year. His knowledge and the diversity of his work in geophysics made it possible for him to be in direct contact with the development of all IVY disciplines, but particularly with aeronomy. Together we saw aeronomy becoming more and more an experimental science, with the peaceful use made of rockets, and during the IVY (1957-58), with the launching of the first Sputniks, Explorers, and Vanguards.
In 1955, Chapman offered the only explanation yet proposed for the rare occurrence of meteor trails that endure for an hour or more, and ten years later, in 1965, with Kendall, he proposed a theory of the conditions for the appearance of noctilucent clouds. Today, Sydney Chapman continues to be interested in agronomy and he teaches the subject in graduate courses at the University of Alaska.