200 Mr. G. A. Schott on the Elect 



ron 



Each of the D lines is numbered 10, and their energy may 

 for our purpose be taken as a measure of that of every other 

 line 10. 



Eder and Valenta"* give the times of exposure required to 

 produce distinct photographic impressions under the same 

 conditions of certain ultraviolet lines : the following are 

 examples : — 



Mg lines, X=2936, 2928, 2802, all 10, 15 sec. 

 Mg lines, X=3336-3329, 2782, all 6, 90 sec, 

 Zn lines, A, = 2063 -2024, all 1, 1200 sec. 



Assuming the photographic effect to be proportional to the 

 product of intensity by time of exposure, we see that the in- 

 tensities are as 1:1/6: 1/80, practically as 1 : (1*6)- 4 : (1'&)~K 



Kayser f proposes to use a scale due to Jewell in which 

 the intensities are as 1 : 2~ 4 : 2~ 9 . We shall not be far wrong 

 if we take as our ratio, for 1 and 10, 1 : 100. This makes the 

 energy of the weakest observable line, 1, 10~ 14 erg per sec. 

 per atom, the radiation in the 1) line being taken as 4 . 10 -13 



(§ 6). 



In all probability this energy is underestimated, probably 

 very much so for the ultraviolet ; PfliigerJ, from an investi- 

 gation on the relative intensities of spectrum-lines by means 

 of a thermopile, concludes that the extreme ultraviolet lines 

 are enormously stronger than the visible lines, although 

 their photographic effect is very much less. For example, he 

 finds that the Mg line, \=280 /jl/jl, gives 970 scale-divisions 

 deflexion, whilst the Mg line \ = 383 fifi, of about the same 

 photometric intensity, gives only 56 scale-divisions; this too 

 although 280 is a single line, whilst 383 includes several 

 neighbouring lines too near to separate. 



§ 11. It is desirable that we should form some idea as to 



the values of o- to be expected for vi- ^ ^^ 



brations of different classes. For this •- — i — ^, 



purpose we shall consider the following jt ^* 



problem : — 1 A 



A ring of n-f 1 equidistant electrons / 

 is in steady orbital motion. An electron T c ' s \ > ^ • 



is suddenly expelled ; required to find the k ^0 ^^"V 



subsequent motion. V J 



To fix the ideas suppose n odd. After ^^v^^^* 

 the expulsion of the electron we are left o 



* Beitrdge zur Photochemie, p. 44. 

 t Spectroscopic, vol. i. p. 646. 

 | Drude, Annalen, xiii. p. 890. 



