PHENOMENA RELATING TO THE SPECTRA OF HYDROGEN AND HELIUM. 277 
precision, although such lines cannot be seen as resolved on account of their width. 
Even when the correction involving /3 is not applied, so that account is not taken of 
the variation of dispersion within the small breadth of the line, the result is 
o 
SX = 0'039 A.U. This would be sufficient to indicate that there was no Diffuse series 
component, and it gives an estimate of the comparatively small degree of dependence 
of the result on irregularity of dispersion. 
(XIX.) Discussion. 
We may at once state that we have at present no theory to offer for the results 
obtained, but it would appear that the changes in relative intensity which we have 
found are facts which must be seriously considered in theories of the origin of spectra 
in relation to the structure of the atom. It cannot be said that spectroscopy has 
already passed from the descriptive to the rational state. We do not propose to discuss 
in detail the bearing of our results on the application of the quantum theory to the origin 
of spectra, but attention may be drawn to certain considerations. It would appear 
from the well-known relation Ye = hv that the distribution of intensity in a spectrum 
should be a simple function of the distribution of velocity among the electrons in the 
discharge tube, but this view appears to be negatived by the fact that different series 
behave in different ways, and we may probably conclude from this that the range of 
velocities of the electrons is not the sole factor which determines the emission in line 
spectra ; this view appears to be strengthened by such phenomena as the apparently 
anomalous effect of Neon on the distribution of intensity in the Balmer series.^ 
As regards the Balmer series, the separations of the components which we have 
found indicate that it is of the Principal type, and lend no direct support to the 
theoretical considerations advanced by Sommerfeld. In any case it can no longer be 
regarded as a Diffuse or Sharp series. 
We do not venture to speculate further on the results we have obtained, which can 
only be regarded as preliminary in their scope, but we hope that the methods of 
investigation which we have outlined may lead to the detailed knowledge which must 
be essential to the further elucidation of spectrum series. 
(XX.) Summary. 
1. A method has been found for the accurate determination of the photographic 
intensities of spectrum lines, and the reduction of such intensities to absolute values 
by comparison with the continuous black body radiation of the Carbon arc. 
* The remarks already made in a previous communication ( loc. cit.) as to the sources of error attaching 
to photographic estimates of intensity appear to be applicable to some results obtained by Rau (‘ Sitz. 
Phys. Med. Ges. Wurtzburg,’ February, 1914), whose conclusions appear to imply that a critical voltage, 
related to the frequency, is necessary to produce spectrum lines. 
2 Q 
VOL. CCXVII.—A. 
