﻿

and the Theory of Atomic Structure. 565 



reached for the rare-gas Xe containing a subgroup of four 

 5 1 electrons and a subgroup of: four 5 3 electrons. For the 

 other regions of the periodic system, however, where, accord- 

 ing to Bohr, an inner shell of electrons is being completed 

 (in the neighbourhood of the iron group, of the palladium 

 group, and of the platinum group) no other shell with higher 

 total quantum number than the shell in question has reached 

 a stage of completion. 



§ 4. The new measurements of Siegbahn and Dolejsek in 

 the K-series* show that for the elements with lower atomic 

 number than 18 there seems to be some irregularity in the 

 relation of the frequency of the line Kft to the atomic 

 number. This irregularity has been connected with the 

 development of the M-shell. For the elements in the neigh- 

 bourhood of the iron group, however, where the second stage 

 of this development takes place, no such irregularity was 

 observed. It could perhaps be supposed that there should 

 exist such irregularities for the lines L/3 2 , I^, Iyy 2 , and L73 

 in the region of the rare-earths, and that the appearance of 

 the satellites on the long wave-length side of these lines 

 should be connected with this fact. Extrapolating these 

 lines, however, from the values for the elements which either 

 precede or follow the rare-earths in the periodic table, we 

 get in both cases values which agree fairly well with those 

 of the measured lines, and at any rate agree much better 

 with those than with the values for the satellites. If there 

 are some irregularities in this region — as we should expect 

 from what has been said about the binding energy of the 

 4 3 -electrons (Part III. § 8) — they must lie within the 

 limits of experimental error. In this connexion, however, 

 it would be of interest to measure the lines @ 2 , 71, 725 and 

 7;. for the rare-earth metals and for the elements in the 

 neighbourhood of this group with the method of high pre- 

 cision recently developed by Prof. Siegbahn. Unfortunately, 

 this would be very expensive work. 



§ 5. As has already been stated in Part III. § 8, it may 

 be assumed that the remarkable satellites, /3 14 , 79, and 7,0, 

 dealt with in Part IV. § 11, are connected with the comple- 

 tion of the inner N-shell. A possible explanation of these 

 lines is suggested in Part V. § 9. Here I shall only con- 

 sider certain special points. Though the satellites on the 

 short wave-length side of /3 2 and y Y (i. e., /3 n and /3 12 and 77) 

 are very faint for the region Ba-Lu, and though they are 

 partly overlapped by other lines, we may conclude, from a 

 thorough examination of the plates, that there is not the 



* Siegbahn and Dolejsek, loc. cit. 



