Jack — Magnetic Resolution of the Spectrum Lines of Niobium. 47 



through lack of intensity. It will be seen that one must therefore assume 

 here that many components have disappeared. For the second line the 

 intensities are small, and hence it is quite possible that the number of 

 components wanting might thus be explained. 



The following three lines have each 8 components. 

 Table II. 



A = 3254-23 



A = 4163-81 



A = 4289-62 



i 





A 



i 



A' 



A 



i 



-^ i A 

 a' 1 



2 



15 

 2 

 3 



± 1-71 s 

 ± 1-35 p 

 + 1-08 s 

 + 0-41 s 



13x-I34(^) 

 10 



8 



3 



5 



8 

 8 

 5 



± 2-40 8 

 ± 1-56 a 

 ± 1-26 p 

 ± 0-71 s 



40x-060(^) 



26 



21 



12 



3 

 1 

 2 

 1 



± 1-98 s 

 ±'l-07 p 

 ± 0-89 s 

 ± 0-36 p 



11.-179 g) 

 6 

 5 



2 



There is no similarity between any of these. Again, it is seen from the 

 A- columns that the Eunge rule is applicable. It might be objected that this 

 is not the case for the second line, since the aliquot part zr^ is required. 



lo 



However, it must be noted that the components of this line are very sharp, 

 and the intensities are also favourable to a very accurate measurement of the 

 separations. The intervals calculated from these observations agree so well 



with the stated multiples of this aliquot part ( — J that, taken in conjunction 



with the possibility of accurate readings, there can be no hesitation in 

 advancing the case of this line as for, rather than against, Eunge's i-ule. If 

 these lines be taken as examples for the application of the Eitz law, then the 

 following values are obtained : — 



\ = 3254-23 

 = 0-67 \ 111 = 2 

 ' = 1-08 ) n = 0, 1 



A = 4163-81 

 = 0-42 I m = 3 

 = 1-56 I w = 0, 2 



X = 4289-62 

 I) = 0-36 \ m = 1, 3 

 = 0-89 ) w = 0, 3 



Thus for all these lines satisfactory agreement is obtained for both of 



5a 

 those rules. The frequent repetition of the interval -— between the components 



o 



of tlie first line will be observed. 



