444 Mr. 11. Phillips on the Use of Chemical Symbols. 



been felt by others, I shall offer a few remarks on the memoir 

 above alluded to. 



In page 255, (Phil. Trans, for 1833, Part II.) Mr. Graham 

 states, "that common phosphate of soda is a phosphate of soda 



and of water, and its symbol is Na 2 H P." Here we have 

 the atomic constitution of the salt symbolically expressed ; but 

 neither its analysis, atomic weight, the atomic weight of its 

 acid, alkali or water, is given, nor is there any reference to 

 the author of the system of notation adopted. It became re- 

 quisite, therefore, in order to understand the experiments 

 detailed in the paper, to ascertain the nature of the salt in 

 question by deciphering its formula. The notation used by 

 Mr. Graham I supposed to be that of Berzelius; I therefore 

 referred to his Essai sur la Theorie des Proportions Chimiques, 

 and there found phosphas natricus cum aqua represented by 



Na P-f 24 Aq. It is composed of 



1 atom of soda 781*84 



1 atom of phosphoric acid... 892-30 



24 atoms of water 2698*46 



4372-60 

 Berzelius (Traite de Chimie, torn. v. table, p. 16,) now re- 

 presents phosphate sodique, by Na 2 P +24 H. It is com- 

 posed of 



2 atoms of soda 390-897 x2 = 781*794 



1 atom of phosphoric add 892*285 



to which add 24 atoms of water 2698*460 



4372*539 

 Now, although it is evident that Berzelius's views of the 

 atomic constitution of this salt must have altered, its compo- 

 sition remains unchanged : let us, then, examine whether we 

 can trace this salt (neglecting the exact proportion of water) 

 in the symbols used by Mr. Graham. When Berzelius repre- 

 sents soda by N a 2 , J? is the symbol of phosphoric acid, but 



Mr. Graham gives P; when, on the other hand, Berzelius de- 

 signates soda by N a, the phosphoric acid is P, which Mr. 

 Graham adopts with N a 2 . Berzelius represents water either 

 by Aq, which was his former method, or by H which is his 



present symbol ; but Mr. Graham gives us H, which in Ber- 

 zelius's original plan had no place, and in his recent one, sig- 



