Separation of the Spectral Lines of Thorium 7 



components were too entangled and bewildering to be of any 

 service, and the plates were not saved. This resulted in some 

 loss in the comparison of the intensities of the s- and /'-compo- 

 nents. The relative intensities of the s- and ^-exposures can there- 

 fore only be roughly approximated. To obtain an accurate basis 

 of comparison, the ^-intensities may need to be reduced about 25 

 per cent. The intensities here recorded are multiples of apparent 

 brightness of the components in the first order spectrum, taking 

 a line which is just sufficiently defined to be continuous latterly 

 across the plate as unit intensity. 



In the dissymmetry theory there is a dissymmetry in the inten- 

 sity of the components. The component which has a smaller 

 displacement than its companion should have greater intensity. 

 In Mr. Jack's lines reported by Zeeman (/. c.) there is no such 

 inequality recorded. In thorium the least displaced component 

 is always the stronger. There are a few apparent exceptions, 

 but these are invariably due to overlapping components of adja- 

 cent lines. However, there are a large number of lines which 

 show this dissymmetry in intensity for which there was no dis- 

 symmetry in separation. Some of these, again, are plainly over- 

 laps, but there are a great many which are not. Farther, there 

 are cases where, components are broader than their conjugate 

 components. No accurate record could be made of this, for the 

 effect might be slight or large. Sometimes it was the red, some- 

 times the blue component which was widened. It seemed pos- 

 sible that one might have here components of two very close lines 

 whose separations differed but little in magnitude. The width 

 of some of the undisturbed ^-components sometimes suggested 

 the same thing. But the latter might be close doublets which 

 would separate with stronger fields. The no-field plates actually 

 showed some of these to be two lines. A no-field line shaded to 

 one side might suggest the presence of other lines also, but this 

 is too common an occurrence with spectral lines to be definite. 

 If one could omit all of these there were still lines which showed 

 this action for sharp no-field lines. It might appear that the 

 broadened line was going to separate into two lines with stronger 

 field, leaving an unequal number of components on the two sides 



95 



