Normal State of the Helium Atom. 



127 



in making a similar test of the Lande model. This formula 

 gives approximate values of the frequencies of the various 

 series terms when 8 is given the proper constant value, and 

 m various integral values. Conversely, if the observed 

 frequencies of the various terms are inserted in the left-hand 

 member, the equation can be used to compute a series of 

 values of 8 which are very nearly equal and vary in a con- 

 tinuous manner from one term to another. 



The term for the normal state can be computed from the 

 ionization potential by means of the relation 



Ye = hv, 



(2) 



or from the wave-length of the ultra-violet helium line just 

 discovered by Fricke and Lyman on the basis of the assumption 

 that this line is produced by an atom dropping from the 

 stationary state (2, S) to the normal state. The two methods 

 of calculating the value of this term, which we will denote 

 by (N), are in agreement, and yield the value 203,000*. 



Fig. 2 shows 8 plotted as a function of m, assuming that 

 (N) is identical with ( 1 , 8) . The curve is clearly discontinuous 

 at the point m = l, and shows that the normal state is not a 

 member of the suggested series of stationary states. 



Fig. 2. 



0.3 



0.2 



c 



>(N) 











1 



*" — -c 



> — - ~< 



> ( 



> ( 



> - - 





Te 



rm Nl 



imber ■ 



rr\ 





0.0 



I 2 3 H 5 



The third objection to the model proposed by Franck and 

 Reiche is that it does not harmonize with the spectroscopic 

 observations of Fricke and Lyman. Experience shows that 

 under ordinary conditions electron-jumps between orbits of 

 the same system whose angular momenta differ by ///lV arc 



* To obtain perfect agreement between these two observations of the 

 energy of the normal state, the rather high value of/* obtained from the 

 study of the excitation potentials of other spectrum lines must be used. 

 Cf. Franck & Einsporn, Zeitschr. fiir Fki/s. ii. p. 18 (1920). 



