PART SECOND. 



The Absolute Atomic Weight of 

 Ten Leading Elements. 



\. OUR METHOD OF DETERMINATION. 



Having briefly presented the mathematical and chemical 

 methods equally required in the making of any atomic 

 weight determination, and having indicated some of the 

 common errors committed in work of this kind, we may 

 begin the exposition of our own method by which we have 

 obtained the absolute values of the atomic weights of the 

 chemical elements. 



It has been demonstrated that the common habit of 

 arbitrarily (i adopting " some set of atomic weights in the 

 reduction of new determinations is not only absolutely irra- 

 tional in theory, but leads to gross errors in practice. 



While chemists have been calculating their new work to 

 the second and third decimal, they have, by the above prac- 

 tice, started out with errors ten and even a hundred times as 

 large. See pp. 33-37. 



Absolutely Fixed Points Needed. 



As in triangulation and even in common leveling, per- 

 fectly well marked starting or base points are required and 

 carefully made often at great labor and expense, so ive must 

 in this ftmdamental ivork in chemistry use certain absolutely 

 fixed data in all our calculations, in order to avoid the intro- 

 duction of errors by the process of calculation or reduction. 



Now such data we have in the indisputable fact that the 

 atomic weights of all elements are quite near whole numbers, 

 if we take oxygen as 16 exactly. 



