374 



90 



other hand a very strong component has been observed in the spark discharge image, 

 which lies in the middle between the theoretical positions of (13 -* 21) and (04 — 21). 

 On the present theory it seems necessary to assume that the component VII corre- 

 sponds to (IB-' 21) only. In fact, the intensity of the component (04 — 21) should 

 be expected to be connected with the value of the amplitude of the vibration of 

 frequency ( — 2oj^ ^ 'åw^) in the motion of the electron, but under the influence of 

 a perturbing homogeneous electric field such frequencies do not in first approxima- 

 tion appear in the motion of the electron in the atom. The component in question 

 is much stronger than the other two separate new components (31-^21) and 

 (31^03), in agreement with the table according to which (13^21) is the strongest 

 new component after (13 03). On the other hand it remains a remarkable fact 

 that the new component IV appears so much stronger than the original component 

 III which lies quite near to it, while it does not appear at all in the continuous 

 discharge spectrogram. Apart from this difficulty, which perhaps will disappear 

 when further experimental data become available, the observations on the fine 

 structure of 4686 A seem to allow of a complete theoretical interpretation if the 

 effect of the presence of electric fields in the vacuum tube is taken into account 

 in the way described in the preceding. As regards the intensities of the perturbing 

 electric forces in question it is seen from the table, which gives the values of Ihe 

 relative amplitudes for a force of 300 Volt cM, that forces of 500 à 1000 VoU/cM would 

 be sufficient to give rise to new components of considerable intensities which are 

 of the same order of magnitude as those of the new components in the spark 

 discharge image, while in case of the continuous discharge image these forces may 

 have been of the order 100 à 300 VoH/cm. i) 



Let us next proceed to the helium line 3203 À (5 — 3), where the discussion 

 of the observed fine structure is quite analogous to that for 4686 Å. In fig. 11 we 

 have given a scheme of the theoretical fine structure and of the components ob- 

 served by Paschen in the continuous discharge spectrogram. The original compo- 

 nents are again represented by drawn lines, the new components by dotted lines. 

 The lengths of the lines representing the original and the new components are 

 taken proportional to the sums of the quantities appearing in the 5th and 6th 

 columns of table XII, which is arranged in the same way as table XI. The lengths 

 of the new components in the figure correspond lo a perturbing field of 9(1 ^'oit ^.yl, 

 Owing to the factor nl in formula (126), the values for the relative amplitudes in 

 the initial slates, corresponding to the new components, are, for same values of F, 

 several times larger than for the new components in 4686 A. This seems lo offer an 

 explanation of the fact, mentioned by Paschen, that in the spark discharge spectro- 

 gram of the fine structure under consideration no sharp components could be 



') In the case of one of the pliotograplis |>ublislied by Haschen, wliich is taken under such condi- 

 tions that the electric current had to pass a long sparking length, the electric forces have obvioush 

 been very large since the two strongest components could not be observed separately but appeared 

 as a broad diffuse line. 



