of Notation in Chemistry.? m 449 



by repeating the second letter of the symbol for each additional 

 atom of oxygen, and adding s (semissis) for half an atom. 

 Thus Mn, Mns, Mrw are the protoxide, deutoxide, and pe- 

 roxide of manganese. 



In the notation of Berzelius, the atoms of oxygen are indi- 

 cated by dots placed over the symbol of the base. Thus, 



fe, fe are the protoxide and peroxide of iron, which he con- 

 siders as having two and three atoms of oxygen respectively. 

 This notation is compact and simple, but it is not consistent 

 with algebraical rule, so far as the oxygen is concerned ; and I 

 conceive that, if this element be explicitly expressed, it ought 

 to be dorie in the manner I have recommended above, fe+2o, 

 /e+3o, &c. 



The employment of different notations for the two purposes 

 of expressing respectively mineralogical analysis, and the ulte- 

 rior analysis with which chemistry is concerned, has been found 

 so far convenient, that it has been introduced into general use 

 by Berzelius and his followers. They have, however, embar- 

 rassed their method with rules and inventions, which very often 

 make the relation between the two sets of symbols obscure and 

 perplexed. Thus it is by no means, at first sight, obvious that 

 the following pairs of symbols are identical, though the writers 

 of whom I speak make them to be so : 



M 3 Si 4 = MSi*, L Si* = LSi 9 , 

 F Si =/ Si, Fe i si = F s Si. 



The number of atoms of oxygen which we must suppose to 

 combine with the base in given substances is, in some cases, 

 dependent on convention, and in others is not yet accurately 

 determined. Hence, the view taken of many substances by 

 Berzelius and by English chemists is different. Thus, he con- 

 siders silica as a combination of 1 atom of base with 3 of 

 oxygen, and writes it S i. Most English chemists consider it 

 as 1 atom of base and 1 of oxygen = si + o, for which we 

 may write the abbreviation S. 



I will now add the list of both the kinds of symbols which 

 I have recommended, and which I hope the preceding obser- 

 vations have shown to be mathematically consistent and che- 



