Many attempts have been made to detect regularities amongst the numerous lines which constitute the secondary or many-lined spectrum of hydrogen. The extreme complexity of the spectrum may be realised from the fact that in the Bakerian Lecture of 1922 Merton and Barratt record some 750 lines in the interval between Hα (wave-number
v
= 5233.216) and Hβ (
v
= 20564.793). Three methods of investigation may be employed in the search for regularities. (1) The lines may be classified according to their physical characteristics, such as intensity or mode of excitation, as in the tables of Merton and Barrat (
loc. cit
.). (2) Lines may be grouped together by the discovery of relations between their wave-lengths or wave-numbers, as in the important groups of lines which have been arranged in bands by Fulcher. (3) Lastly, the question may be attacked from the theoretical side, and a model of the hydrogen molecule may be imagined, which will give rise to the emission of certain characteristic spectral lines. Thus Sutherland, working on the foundation of the classical mechanical laws, more than twenty years ago, came to the conclusion that spectral series must arise from kinematical considerations, and explained them by considering the nodal sub-divisions of a circle. At the present time we may expect more successful results to follow from the application of the quantum theory, and in this paper an endeavour will be made to examine the secondary spectrum of hydrogen, and more particularly the Fulcher bands, from this standpoint. I may add that my interest in the subject was aroused when attempting to construct a model of the hydrogen molecule, for it seemed that the most likely method of obtaining reliable information from the experimental side as to the moment of inertia of the molecule would be from a study of the spectrum of molecular hydrogen.
The band spectrum of hydrogen
