i8 ' B. E. Moore 



accurate confirmation of such a relationship. But the multiple 

 relationship is frequently very easily recognized. 



PART III. CONCLUSIONS 



1. Lines which have six or more components in thorium are 

 relatively very few. The separations of a great many of these 

 are multiples of small values. But these small values are not 

 closely related to aliquot parts of a "normal" separation. 



2. There are numerous lines unsymmetrical both in separation 

 and in intensity. The stronger component is always least 

 displaced. 



3. There are a few lines with an unequal number of compo- 

 nents upon the two sides of the zero position. 



4. There are a great many lines unsymmetrical in intensity, 

 which fail to show any dissymmetry in position. 



5. The components are often unsymmetrical in width, i. e. one 

 component is sharp and its opposite companion is broad. Some 

 of the latter look as if they might separate with stronger field and 

 give lines of the type in conclusion 3. 



6. There are numerous lines unseparated in the magnetic field, 

 but very peculiarly affected thereby. Some of them vibrate more 

 strongly perpendicular to the lines of force than parallel thereto, 

 and others vice versa. Some of these are possibly related in pairs. 

 Others may belong in groups. 



7. Numerous triplets have one or more close companions of 

 like separation, which suggest pairs and groups of closely related 

 lines. 



8. There is often a multiple relationship in the magnitude of 

 the separations of unsymmetrical components. This would be 

 merely a coincidence, if the dissymmetry is independent of the 

 field strength or proportional to the square of the same. 



9. The types of separations for lines which have several com- 

 ponents are not repetitions of types which I had found in other 

 substances. Reasons are given for excluding triplets and quad- 

 ruplets from such a comparison. 



106 



