Separation of the Spectral Lines of Thorium n 



choose lines corresponding to either of these ^-separations, one 

 would find the />-components to have as great a variety in their 

 separations. There are a number of the ^-separations reasonably 

 near to the value a, but very few, however, of the corresponding 

 ^-components come near to aliquot parts of a. These relations 

 suggest that there is no particular meaning to be attached to a 

 "normal" separation among these quadruplets. None of these sep- 

 arations approach reasonably near to the quadruplet principal 

 series, neither were there representatives of the companion sex- 

 tuplet series. There are lines having the .?- and other lines having 

 the /'-separation of the same magnitude as the quadruplet prin- 

 cipal series, but no lines having both components coincidentally. 



Table XI contains a list of sixty />-doublets. The intensities 

 follow the order of Table X. The weakest of these lines do not 

 appear upon the ^-plates, and only upon one set of /opiates. On 

 account of overlapping, it was impossible to make out the char- 

 acter of the separations for some ^-components, and impossible 

 to measure others. Some ^-components were too diffuse to meas- 

 ure. It is safe to say that most of these lines are quadruplets. 

 On account of the wide variation in the magnitude of these ^-sep- 

 arations it would be no easier to arrange and classify them than 

 a similar number of triplets. If one may add the greater portion 

 of them to the quadruplets, they increase the already large vari- 

 ation of these components and make it more improbable that there 

 is any standard in the magnitudes of these separations. 



Tables XII-XVI inclusive contain a list of unsymmetrical lines 

 exclusive of triplets. 



Table XII is an unsymmetrical 9-component line. There are 

 two red and one blue /'-components. The latter is double the in- 

 side red component. The remaining red component has the value 

 of the "normal" triplet, but it is not related in a simple way to the 

 other /'-separations, nor to the separations of the ^-components. 



Table XIII is an unsymmetrical 7-component line. Both p- 

 and .y- are unsymmetrical ; but the strong undisplaced /'-compo- 

 nent is the only one which can be measured with desirable 

 accuracy. 



Table XIV gives four 6-component lines. The footnotes indi- 

 cate their principal features. 



99 



