Normal State of the Helium Atom. 131 



must be remitted at low pressures without change in wave- 

 length. Franck and Reiche base their argument for the 

 metastable character of the state (2, s) on Paschen's obser- 

 vation that the line at 10830 A.U. does actually have bids 

 property. They further make the statement that the line at 

 20582 A.U. is not a resonance line ; but this seems to be a 

 mistake, for the writer finds no mention in Paschen's paper 

 of any investigation of the intensity of the fluorescence of the 

 latter line. 



The Angular Momentum of the Normal Helium Atom. 



In discussions of atomic models it is commonly assumed, 

 though without general proof, so far as the present writer is 

 aware, that the average angular momentum of each electron 

 and the resultant angular momentum of the entire atom are 

 integral multiples of h\2ir. These hypotheses are at least very 

 plausible, and are assumed in the discussion which follows. 



The Bohr 'and Lande models of the normal helium atom 

 have each two units of angular momentum*. The work of 

 Lande indicates further that the orbits of the stationary 

 states of the single-line system are so inclined that the 

 resultant angular momentum of the atom is equal to the 

 mean angular momentum of the outer orbit alone. It follows 

 that the helium model of Franck and Reiche has just one 

 unit of angular momentum. On the other hand, the simplest 

 explanation of hypothesis (a) is that the normal helium atom 

 has zero angular momentum. 



The Bohr principle of selection requires that when an 

 atom undisturbed by external forces jumps from one stationary 

 state to another of the same family, it shall radiate one, and 

 only one, unit of angular momentum. If this principle is 

 assumed to apply also to jumps from a stationary state 

 of one kind to another of a wholly different character, one 

 may draw from it the conclusion that the normal helium 

 atom differs in angular momentum by one unit from the atom 

 in the state (2, 8). Since the latter, like the model of the 

 normal atom proposed by Franck and Reiche, has one unit, 

 the normal atom must have two units or none ; but if the 

 average angular momentum of each electron is one unit, the 

 only way in which the resultant angular momentum can be 

 two units is for the electrons to execute coplanar orbital 

 motions in the same direction. This possibility is ruled out 

 by the fact that the Bohr and Lande models, to which it 

 leads, do not have the right energy, and also by the absence 

 * Lande assumes that the electrons rotate in the same direction, re- 

 marking in a footnote that " Komplanare Gegenrotation wurde das "\ or- 

 zeiehen der Storungen lOrdnung umkekren, eutgegen der Beobachtung, 



K2 



