Linus Carl Pauling was the firstborn, in 1901, of a pharmacist in Portland, Oregon, who died in 1911 leaving his wife, son, and two daughters with limited means. After high school in Portland, Linus Pauling entered Oregon Agricultural College at Corvallis, precursor of Oregon State University, in 1917, and graduated in chemical engineering in 1922. He worked his way through college, serving as full-time assistant instructor in quantitative analysis 1919-1920. The experience may have dissuaded him from accepting a half-time instructor's post for five years of graduate study for a PhD at Harvard. Instead he moved, in 1922, to a three-year graduate studentship offered by Arthur Amos Noyes (1866-1936), head of the Division of Chemistry and Chemical Engineering in the California Institute of Technology (Caltech) at Pasadena.

Noyes had an eye for talent and for promising new fields of research, and it is said that Pauling was Noyes' greatest discovery. Noyes obtained his PhD with Wilhelm Ostwald (1853-1932) at Leipzig in 1890, then joined the Massachusetts Institute of Technology (MIT) where, as professor of theoretical chemistry 1899-1919, he recruited a number of able younger chemists. These included Gilbert Newton Lewis (1875-1946), who was at MIT 1908-1912 before moving to the University of California, Berkeley, as head of the chemistry department. Noyes commuted to Pasadena each winter from 1915 to build up the chemistry division of Throop College of Technology, which changed its name to Caltech shortly after Noyes moved permanently to Pasadena in 1919.

Noyes recognised the importance of X-ray crystal structure analysis from the beginning, and installed X-ray equipment at MIT and Caltech. Roscoe Gilkey Dickinson (1894-1945) was in charge of the powder and single-crystal X-ray apparatus at Caltech in 1922 when Linus Pauling was placed with him by Noyes for research supervision as a graduate student. Dickinson and Pauling published their first paper in 1923, on the structure of the mineral molybdenite, MoS₂ establishing a trigonal prismatic coordination of molybdenum by six sulfide ions. Pauling soon achieved scientific standing, as author or coauthor of about a dozen crystal-structure publications over the next three years, and G. N. Lewis offered him a postdoctoral position at Berkeley after his PhD in 1925. Noyes thereupon arranged a Guggenheim fellowship for Pauling's postdoctoral studies in Europe 1926-1927, centred on the Munich Institute of Arnold Sommerfeld (1868-1951), indicating that a position at Caltech would be available on Pauling's return.⁹

In Europe for nineteen months, 1926-1927. Pauling met the principal workers in the field of quantum mechanics as they came to visit Sommerfeld's Institute at Munich, or on his own visits to Copenhagen and Gottingen for a few weeks, and to Zürich for several months. As a graduate student, Pauling, had attended a wide range of advanced courses on mathematics and the physical sciences, and soon assimilated the concepts and procedures of the new quantum mechanics. He said later on that he did not bother overmuch with the deeper philosophical implications of the uncertainty principle and the like. Following the pragmatic tradition of North America, Pauling adopted an operational approach to the new discipline, seeking concrete applications of quantum mechanics to chemical and physical problems.

At Bohr's Institute in Copenhagen Pauling met Samuel Goudsmit (1902-1978) who, with George Uhlenbeck (1900-1988), introduced in 1925 the physical notion of electron spin to account for the two-valued fourth quantum number needed in atomic spectroscopy. The new number had entered empirically into Pauli's principle (1925), forbidding the same set of four quantum numbers to any two electrons in any given polyelectronic system. Pauling and Goudsmit later collaborated in writing The Structure of Line Spectra (1930). More momentous was Pauling's visit to Schrodinger's Institute in Zürich, where he met Fritz London (1900-1954) and Walter Heitler (1904-1981), who were working on their valence bond (VB) treatment of the bonding in the hydrogen molecule, published in 1927. The two electrons (1) and (2) of the molecule are allocated to the ls atomic orbital around each nucleus, H_a and H_b in two ways, [H_a(1)H_b(2)] and [H_b(1)H_a(2)], to give two 'valence structures'. Calculations indicated that, at bonding internuclear separations, the principal source of the molecular binding came from the 'exchange energy', arising from the interchange of the two electrons, with opposed spins, between the two 'valence structures'.

About the same time Friedrich Hund (b. 1896) developed the alternative molecular orbital (MO) treatment of the bonding in the hydrogen molecule at Gottingen. On the MO model the paired electrons move in a molecular orbital resulting from the in-phase combination of the 1 s atomic orbitals of the two nuclei, [H_a+H_b]. Subsequent comparisons of the two methods showed that the original MO treatment gave ionic structures of the type [H_a(1,2)] and [H_b(1,2)], additional to the neutral valence structures of the first VB treatment, and of equal weight. The two methods became identical, and gave a theoretical bond distance and bond energy closer to the corresponding spectroscopically measured values, when the weights of the contributions from the ionic structures were reduced in the MO treatment and were added to an equivalent degree in the VB treatment. The conceptual differences between the VB and the MO methods remained, however, in the simplified and approximate methods needed for the treatment of complex polyatomic molecules. These differences occasioned some contention between advocates of the VB and the MO methods until the 1950s, when the growth of chemical spectroscopy brought about the general adoption of the MO procedure, with its more fruitful treatment of excited molecular states.

In North America the principal advocate of the MO theory was Robert Sanderson Mulliken (1896-1986), at the University of Chicago from 1928. Mulliken was a close friend of Hund from the mid-1920s, and regretted that his Nobel Prize (1966) was not shared with Hund.¹⁰ During the pre-war period, chemists took little note of the MO studies of Hund and Mulliken. The early MO models regarded a molecule as a fixed array of atomic nuclei, each with its own completed inner shells of electrons, while the electrons of the incomplete outer shells of the atoms, the 'valence electrons', moved in molecular orbitals spanning the array of atoms as a whole. There were no individual 'chemical bonds' in a polyatomic molecule, according to early MO theory, contrary to classical structural theory. Traditionally, chemists constructed molecules, conceptually and in the laboratory, by adding another atom or group, through a well-defined 'chemical bond', to a simpler structure.

Mulliken opened his Chemical Review of 1931 with the opinion that 'the concept of valence itself is one which should not be held too sacred'. After devoting a section to the 'Superfluity of the concept of valence bonds in the 'molecular' point of view', he came to the conclusion that the VB method, 'when applicable, usually gives, somewhat fortuitously in the author's opinion. the same results as the present [MO] method. The latter gives, however, a detailed insight into what is going on in the formation of the molecule.¹¹ During the 1930s few chemists accepted Mulliken's views of chemical bonding. In contrast, Pauling's resonance theory, formally based on the VB method, aroused widespread interest, particularly in North America, since it preserved and rationalised much of classical structural theory and the pre-quantum mechanical theories of the role of electrons in chemical bonding, developed mainly by chemists.

In 1927 Pauling returned to Caltech as assistant professor in theoretical chemistry, and began a series of investigations on the nature of the chemical bond, alongside his resumed X-ray studies of crystal structures. In 1930 he extended his structural studies to individual molecules in the gas phase, free from the complexities of the packing of molecules in crystals, with the new technique of electron-diffraction, developed by Hermann Mark in Ludwigshafen. Pauling visited Mark early in 1930 when he spent some time with William Lawrence Bragg (1890-1971) at Manchester. With Bragg he discussed various crystallographic procedures, including the applications of Pauling's rules (1928) governing the geometry of the coordination polyhedron of anions around a cation in an ionic crystal, in terms of the radius ratio of the anion and the cation, and their formal charges. These rules were elaborations of rules proposed 1923-1926 by the geochemist crystallographer, Victor Moritz Goldschmidt (1888-1947) in Oslo, and they had particular value for the structural analysis of the silicate minerals, which Bragg and Pauling were studying.

Pauling recalled in 1991 that his interest in electronic theories of chemical bonding dated from the time he served as assistant instructor 1919-1920. One of the two chemistry seminars that year at the Oregon Agricultural College was given by an agricultural chemist on the frozen fish industry, while Pauling spoke on the shared electron-pair chemical bond. This basic idea had been proposed by G. N. Lewis in 1916 and developed in a series of papers from 1919 by Irving Langmuir (1881-1957), who coined the terms 'covalance' and 'electrovalence' for the homopolar and the heteropolar sharing. The Coulombic attraction of opposite charges provided a physical basis for the electrovalent (ionic) bond, but the homopolar shared-pair covalent bond had no immediate physical foundation, other than the significant correlation with the electron-pair of the lightest noble gas, helium, and the four duplets of the eight electrons in the outer shell of the heavier noble gases, modelling the electron configuration of the central atom in polyatomic systems, such the carbon atom in CH₄.

In Munich and Zürich 1926-1927 Pauling found what he believed to be the physical basis of the homopolar covalent bond in the quantum-mechanical 'exchange energy', arising from the interchange of spin-paired electrons between the two 'valence structures' in the VB treatment of the hydrogen molecule by Heitler and London. Pauling regarded the electron-pair exchange in a chemical bond as the quantum-mechanical analogue of the classical resonance effect observed in coupled oscillators, terming the bond energy from electron interchange the 'resonance energy'. He referred the analogy back to the 1926 treatment by Werner Heisenberg (1901-1976) of the separate para- and ortho-states of the helium atom (spin singlets and triplets, respectively), which resembled a classical case of the resonance splitting between the in-phase and out-of-phase modes of coupled oscillators. Pauling introduced his resonance theory in a 1928 Chemical Review and developed his ideas in a series of seven papers 1931-1933 on The Nature of the Chemical Bond, culminating in his George Fisher Baker Lectures at Cornell University, 1937-1938. The lectures were published, The Nature of the Chemical Bond in 1939, with a second edition in 1940 and a third in 1960. All were dedicated to G. N. Lewis, whom Pauling regarded as the founder of the modern theory of valence.