348 



64 



nualogy with what was the case for Hy and //,;, that this estimate will he exagger- 

 ated in the case of components corresponding to transitions for which the electron in 

 the final states describes a circular orbit (J = 7, J = 21; J = 0, J = 14), in 

 such a way that these components will appear stronger than it would be expected 

 from the values of R' ^ in Table V. 



The considerations in this chapter may naturally also be applied to the pro- 

 blem of the Stark effect of the lines of the helium spectrum which correspond to 

 iV = 2 in (105). The experimental data for these lines, however, are not nearly so 

 complete as for the hydrogen lines, only a few measurements referring to the strong 

 visible line 4686 Å (4 3) and to the ultra-violet lines 3203 Å (5 ^ 3) and 2733 Å 

 (6-^3) having been published. When a strong electric field is applied we must 

 expect that the lines in question will show a symmetrical resolution in a number 

 of components the frequencies of which can be obtained from (HI), and an estimate 

 of the relative intensities of which can be obtained from the preceding considera- 

 tions. Table VI contains the values of R' '^ and R" ^ corresponding to the components 

 of the Stark effect of 4686 A, while fig. 7 on Plate III contains the schematical pic- 

 lure of the theoretical intensities. An observation on the Stark efïect of the 4686 Å 

 line in helium for a comparatively small electric field has been published by Evans 

 and Croxson 1) and is also contained in a recent paper by Nyquist-') on the effect of 

 an electric field on the helium spectrum. The photographs of both these authors show 

 distinctly that the resolution of the line in question is symmetrical, but the electric 

 tield used in their experiments was not strong enough to separate the different 

 components the existence of which is claimed by the theory. Thus Evans and Croxson 

 observed only a symmetrical broadening of the line, but in Nyquist's photographs 

 the line in question was resolved into an undisplaced perpendicular component 

 and two symmetrical parallel components, which are indicated in fig. 7 by arrows. 

 The distance of each of the latter from the undisplaced line amounted to 3,2 A for a 

 field of 100.000 Volt/cM (the largest intensity of the field in the experiments amounted 

 to 38600 Volt/cM). This would correspond to a value of J equal to ^,2 0,71 = 4,5 

 which value is seen to be in excellent agreement with the position of the centre of 

 gravity to be expected for the strong theoretical parallel components at J = 2, 3, 4, 

 5 and 6 (compare Table VI and fig. 7). 



Measurements on the effect of an electric field on the ultra violet helium 

 lines 3203 Å (5 -^3) and 2733 Å (6-3) have recently been published by Stark«). 

 Also in these experiments the intensity of the electric field, which amounted to 

 28500 Volt/cM, was not strong enough to obtain separately the different theoretical 

 components. Tables VII and VIII, which are arranged in the same way as the pre- 

 ceding tables, contain the values of R' ■ and R" - corresponding to these components, 

 while fig. 8 and 9 on plate 111 contain a schematical picture of the results obtained in 



') E. J. Evans and C. Ckoxson, Phil. Mag. XXXII, p. ;i27 (lOKi). 

 -) H. Nyqui.st, Phys. Rev. X, p. 226 (1917). 

 ^) .1. Stark, Ann. d. Phys. LVI, p. .509 (1918). 



