16 Prof. W. M. Hicks on the 



Y 1 .e. F 2 .e. 



(2m) 17269-10=e+15 



(4) 17050-97 arc. 



• 







208 



64 84 







[17071-8+.r] 



[17319-8-f.r'] 



{ 2424 

 1 2541 



(In) 17562-23 

 (3) 17573-91 XL 0O 



2621 



{In) 17333-94 

 1579 









(In) 17349-73 



2512 



(1^)17601-92 





623 









(5) 17355-96 







It will be noticed, the same weak and nebulous character 

 of the lines, the frequency of separations depending on 

 multiples of 4*9, and also similar analogies in the two 

 sets, e.g+the changes from a sharp line to a nebulous one 

 with sep. 11'6 or in that neighbourhood at 17456, 17573. 

 In relation to the whole set it should be remembered that we 

 should expect the 27 $ { or 2SB X displacements of the D series 

 to be repeated on the limit, or on the /sequent. In this 

 connexion the following data may be given : — 



+ 27^ -278, +288, -288, 

 248-91 252-39 258'05 261-80 



The effects are capable of so many interpretations that 

 it is impossible to draw any certain conclusions. A few 

 remarks, however, may not be out of place : — 



(1) f(l) is taken as <i 9 (l)(76\), so that the set are 

 ' d 1 (l)±f(l) andd 2 (l)±/(l). 



(2) 17178 is either *.(oY)F 2 or *.F 2 (-o\). 



(3) In 17269 the original displacement of 64'8 is re- 



produced. 



(4) It is difficult to explain the differences 4'9 by oun 



displacements. They could possibly be explained 

 by the oun if these series depended on a Rydberg 

 constant 4N, as in the enhanced series of the 

 alkaline earths, and there are some independent 

 indications pointing to this. 

 (5) The important consequence to be drawn is that they 

 afford evidence of the extreme disruption of a set, 

 and of the nature of the complicated displacement 

 systems called into play and which require dis- 

 entanglement. 



We might now attempt to allocate sets of lines to stand in 

 the relation of higher orders to the series. They exist, but 



