302 



Mr. F. W. Aston 



[Feb. 11, 



but that none of them could be said to carry the conviction neces- 

 sary in such an important development. 



When the work was recommenced, attention was again turned 

 towards positive rays, for it was clear that if an analysis could be 

 made with such accuracy that it could be demonstrated with certainty 

 that neither of the twb atomic weights so determined agreed with 



H 



the accepted chemical figure, the matter could be regarded as settled. 

 This could not be done with tbe parabolas already obtained, but the 

 accuracy of measurement was raised to the required degree by means 

 of the arrangement illustrated in Fig. 1. Positive rays are sorted 

 out into a thin ribbon by means of the two parallel slits S x S 2 , and 

 are then spread into an electric spectrum by means of the charged 

 plates P x P 2 . A portion of this spectrum deflected through an 



