140 BELL SYSTEM TECHNICAL JOURNAL 



of which was until a couple of years ago regarded as perfectly dis- 

 tinct; but at this moment it is beclouded by one of the curious con- 

 tradictions so abundant in the Theory of Atomic Structure. Briefly, 

 the t>'pical phenomenon is this: the differences between the energy- 

 values of the K, Lii and Lm states agree notably well with what 

 would be expected if the complete atom contains a few electrons 

 moving in li orbits, a few in 2i orbits and a few in 22 orbits about the 

 nucleus; and if the K state corresponds to absence of an electron of the 

 first group, the Lu state to absence of one out of the second and the 

 Lin state to absence of one out of the third. (The reason why calcu- 

 lations can be inade for so indefinitely-phrased a model is this, that 

 the field due to the highly charged nucleus of a massive atom should 

 dominate over those of the individual electrons so that it does not 

 make very much difference how many are supposed to be in each 

 group.) The natural inference is, that the rest of the atom remains 

 unchanged or little changed when any one of these electrons is ex- 

 tracted. In this case the Azimuthal Quantum-number of the residue 

 should differ by one unit for the two states Lu and Lm. Consulting 

 Table II, one finds that the cjuantity called k, which obeys the char- 

 acteristic selection rule of the Azimuthal Quantum-number, is the 

 same for Lu as for Lm. This is an illustration of the collisions be- 

 tween two sets of inferences which unsettle the supposedly firmest 

 achievements of this theory. 



Of the theory of molecules, a subject large enough for an article 

 by itself, I can say here nothing more than that it attains some re- 

 markable successes, achieved by and therefore fortifying some of the 

 assumptions made in these pages; notably the assumption that 

 Angular Momentum is a thing required in Nature to assume discrete 

 values spaced at intervals of h/2Tr. 



The final part of this long article has been very unlike the Second 

 Part, in which an atom-model for the atoms of hydrogen and ionized- 

 helium was constructed and endowed with certain fundamental qual- 

 ities, so that it reproduced almost all of the relations of these atoms 

 to radiation with a truly striking fidelity. This Third Part by con- 

 trast has been a thing of shreds and patches. Models for many atoms 

 have been brought forth, but they have not been thoroughly adequate 

 and they have not been concordant with one another. Some were 

 designed with the same fundamental qualities as those given to the 

 model for hydrogen; and scarcely more can be said for any of them, 

 than that it does not positi\ely clash with the properties of the element 

 for which it is devised. Others were made competent to deal with a 



