38 SECTIONAL ADDRESSES. 



With the attainment of a definite conception of the electronic structure 

 of the various atoms it becomes possible to approach the second of the 

 two problems referred to above, namely, the determination of the spectro- 

 scopic terms associated with a given distribution of electron orbits. The 

 first steps were taken by Russell and Saunders,''^- and independently in 

 part by Wentzel,^^ who, in a discussion of the so-called ' anomalous ' 

 terms, which have already been mentioned, made one of the most 

 illuminating contributions to spectroscopy of recent years. To the three 

 ^PP' groups of calcium already known two more were added by Russell 

 and Saunders, who were thus able to show that the five groups formed a 

 series which could be approximately represented by a Ritz formula. The 

 surprising result then appeared that, as referred to the regular triplet 

 limits, the later P' terms were numerically negative. The earlier P' 

 terms, having positive values, certainly originated in the neutral atom, 

 and it could scarcely be doubted that the later terms also had the same 

 origin. 



The existence of these negative terms implies a greater equivalent 

 energy than that required for ionisation of the atom, and it follows that 

 it is possible for an atom to remain neutral while absorbing more energy 

 than that necessary to remove the series electron. Hence, in accordance 

 with a previous suggestion made by Bohr, but unknown to them, Russell 

 and Saunders concluded that the energy must be divided between two 

 (or more) electrons, each of which is displaced to a higher energy level, 

 without the removal of either of them. The detailed numerical evidence 

 led inevitably to the conclusion that both valence electrons might jump 

 at the same time from outer to inner orbits, and that the net loss of energy 

 would then be radiated as a single quantum, i.e. as monochromatic 

 emission. Epstein has shown that such simultaneous transitions of two 

 electrons, with the emission of a single quantum of energy, is con- 

 sistent with the correspondence principle, and similar combinations of 

 transitions into a single emission had already been found to be involved 

 in the theory of band spectra. 



Experimental evidence in support of this view of the origin of the 

 anomalous terms in neutral atoms of the alkaline earths is afforded by 

 the fact that such terms do not appear in the spectra of the alkalis, where 

 the energy required to displace a second electron is known to be much 

 greater than that for the first. Additional evidence is provided by the 

 observation that they do occur in the spectra of most of the elements in 

 which enhanced lines are easily produced. It may be added, in further 

 support of the hypothesis, that PP' groups appear conspicuously in the 

 arc spectrum of silicon, in which lines of the ionised atom are entirely 

 wanting. 



In their theoretical discussion Russell and Saunders showed that the 

 anomalous relations could be explained by assuming that the angular 

 momenta of the two valence electrons are quantised in space with respect 

 to each other, and their resultant quantised with respect to the angular 

 momentum of the residue of the atom. 



Arising out of the work of Russell and Saunders, together with further 



" Phys. Mev., vol. 22, p. 201 (1923) ; Astrophys. Jour., vol. 61, p. 38 (1925). 

 "» Phys. Zeit., vol. 24, p. 106 (1923) ; vol. 25, p. 182 (1924). 



