792 Mr. Ivar Maimer on the 



\/ - is graphically represented as a function o£ tho 

 atomic number N in fig. 3 (as well as the values found by 



Fiar. 3. 



















1 1 













































1 





































































** 

























1 























s 

























1 































































s 













































S 



f 



























































































V 



s* 



































































































































s 



£ 













































^ 



s 











































s, 



& 











































s. 



^ 













































s 



r 













































t* 

































































































13 15 17 19 21 Z5 11 27 Z9 31 33 35 37 39 4/ 43 45 47 49 51 53 55 57 59 61 — * 



Moseley for the elements Al-Zn) . The values are evidently 

 ordered along two right lines — the deviations may be within 

 the limits of the observation errors, which must here be 

 relatively great because of the small incident angles. In 

 the case of Sb, however, the deviation is too great to be 

 explained by observation errors, all the more as the lines on 

 this plate were rather sharp. As is seen from the figure, 

 the agreement is rather good, if the atomic number of Sb is 

 taken = 52 instead of 51. In the L-series measured by 

 Moseley Sb must have the number 51 — accordingly, it seems 

 that such an anomaly is possible that an element has not the 

 same number in all series. 



In the case of lanthanum the lines were rather broad 

 (perhaps the oxide contained barium) and were both 

 followed by a rather strong continuous spectrum. For 

 cerium no lines could further be observed, although the bulb 

 was extremely hard, and the plate was exposed twenty-five 

 times, but only a continuous spectrum, which, however, was 

 a little enfeebled immediately in front of the spot where the 

 a-line ought to have been situated. 



From fig. 3 it is evident that the formula given by Moseley 

 for the a-line of the type 



* = |* (N-1) 2 



