1106 Spectra of X-rays and Theory of Atomic Structure. 



several elements and can be classified according to Moseley's 

 law. In the L-series these are the line /3 8 , which has been 

 found on the short wave-length side of /3 2 for a great number 

 of the elements Ta-U *, and the lines which are dealt with in 

 Part II. §§ 7 and 8 of this paper. In the K-series of the 

 lighter elements similar lines have been found by Hjalmar f . 

 Most of these non-diagram lines seem to disappear again for 

 elements with higher atomic number. Another characteristic 

 feature of these lines is that they always lie on the short 

 wave-length side of a very strong diagram line with which 

 they seem to be connected in some way J. Usually their 

 frequency difference from this line is roughly proportional 

 to the atomic number of the emitting element. 



In order to explain the appearance of the non-diagram 

 lines we might suppose that under certain conditions a sub- 

 group of electrons from which one electron has been removed 

 may give rise to more energy-levels than have been inserted 

 in the diagrams. If this really is the case, these new levels 

 must evidently play a different part from those which appear 

 in the diagrams. As yet they have not been detected in the 

 absorption spectra § : furthermore, they each seem to be 

 connected with one line only and they usually disappear 

 again for atoms of higher atomic number. 



Again, part of these new levels might be attributed to a 

 more complicated excitation of the atom. Suppose that w hen 

 a high-speed /3-particle collides with the atom, under certain 

 conditions two electrons are removed at the same time. If 

 then an electron were to fall back into the atom, this transition 

 would give rise to a new line in the X-ray spectrum which 

 would lie on the short wave-length side of a diagram line 

 connected with the removal of only one electron. In this 

 case we should expect to find a great difference between a 

 characteristic X-ray spectrum excited by the impact of high- 

 speed /3-particles (primary radiation) and a spectrum excited 

 by absorption of X-rays (secondary radiation). For the 

 latter we should expect that no such non-diagram lines 

 would appear. As some of these lines are quite strong, 



* Zeitschrift f. Physik, II. p. 191. 



t Zeitschrift f. Physik, i. p, 439 (1920). 



X The new lines recently detected for the rare-earth metals (see 

 Part III. § 8) are an exception to this rule. 



§ In this connexion it may be of interest to state that the structure 

 found by Hertz (Zeitschr. f. Phys. iii. p. 19 (1920)) for the In- and L 2 - 

 discontinuities seems to be not directly connected with that which I 

 have found for the L-«[ and L- J 8 1 lines respectively. 



