Chemical combination among metals has been studied for a long time. In the early days, chemists had much difficulty in understanding how metals could combine at all among themselves, often in several different proportions, since this often seemed to be at variance with elementary notions of valency. Indeed, in the last century a distinction was proposed between daltonides and berthollides, the former being compounds with simple ratios of the numbers of constituent atoms… consistent with Dalton's atomic theory!… the latter, called after a skeptical French chemist, being the name reserved for compounds with a wide composition range which scarcely seemed explicable on any reasonable atomic model. A few nostalgic souls still employ these distinctions.

An early classic on the subject was a book published by two Italian chemists in 1918. British metallurgist C.H. Desch also wrote a monograph on intermetallic compounds in 1914 (the year when the name “physical metallurgy” was first proposed, and the chemical bias of metallurgy shifted toward a physical one), and he declared that “attempts to form a theory of the constitution of intermetallic compounds… have been comparatively unsuccessful.” When he returned to the subject 20 years later, he felt much more cheerful about the state of the theory because in the meantime Hume-Rothery had introduced the concepts of atomic size factor, electron/atom ratio, and electronegativity (a concept also associated with the name of Linus Pauling) into the discussion of intermetallic phases, or compounds. Today, a battery of sophisticated physical concepts is used to make sense of both binary and multielement intermetallic compounds and they are no longer regarded as chemical anomalies.