194 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1915. 



the wave lengths of the latter can be simply determined. The work 

 of Barkla has shown us that an X radiation, characteristic of each 

 element, is excited under certain conditions when X rays fall upon 

 it. The penetrating poAver of this characteristic radiation increases 

 rapidly with the atomic weight of the radiator. In heavy elements, 

 another type of characteristic radiation makes its appearance. These 

 tAvo types of characteristic radiation have been called by Barkla the 

 K and L radiations, respectively. These radiations can be excited 

 either by X rays of suitable penetrating power or by direct bom- 

 bardment of the element by cathode rays in a vacuum tube. Moseley 

 made a systematic examination of the X-ray spectra of a great 

 majority of the elements. For this purpose the elements examined 

 were bombarded by cathode rays, and the spectrum of the radia- 

 tion examined by reflection from a suitable crystal. He found that 

 the spectra of the K radiation from elements varying in atomic 

 weight from aluminium to silver were all similar in type, consist- 

 ing mainly of two strong lines.^ An example of the spectrum ob- 

 tained for a number of successive elements is shown in plate 5, figure 2. 

 It is seen that with increasing atomic w^eight the Avave length of the 

 corresponding lines diminishes, not irregularly but by definite and 

 well-marked steps. Moseley found that for the K radiation the 

 frequency of the radiation was proportional to (N-a)^, where N was 

 a Avhole number which varied by unity in passing from one element to 

 the next of higher atomic weight and a was a constant about unity. 

 From silver to gold the spectra gi^^en by the L radiations of elements 

 were compared. These spectra consist of about five lines, of which 

 two are relatively very strong. It was found again that the spectra 

 were similar in tj^pe and that the frequency of a given line diminished 

 by definite steps in passing from one element to another. The fre- 

 quency of the radiation in this case was proportional to (N-b)-. 

 where b Avas a constant and N a whole number. Moseley concluded 

 that the value of N in these expressions was the atmmc 7vumher\ i. e., 

 the number of the element arranged in order of increasing atomic 

 Aveight. Taking aluminium as the thirteenth element, he found 

 that succeeding elements Avere expressed by the value of N 14, 15, 16, 

 17, etc., up to 77 for gold. 



There appears to be little doubt that the X-ray spectrum of an 

 element arises from the vibrations of the rings of electrons deep in the 

 atomic structure outside the nucleus. Quite apart from the very in- 

 teresting question of the mode of origin of these very high frequency 

 spectra, it is seen that the fundamental modes of vibration of the 

 distribution of electrons are simply connected Avith the square of a 

 number, which varies by unity in passing from one element to the next. 



1 In later work Rawllnson and Bragg have found that each of these lines is in reality 

 a very close double. 



