1825.3 M.Berzelius's Hypothem of the Atomic Theory, 347 



xjorrectly^ for a single error may wholly pervert the meanino- of a 

 formula, and the consequence of such an error is the more 

 serious, because it cannot, as in common language, be readily 

 detected and corrected by the context. The errors of the press 

 too are more likely to escape notice, and thu« this species of 

 danger, from inaccuracy or inattention, becomes doubled. Much 

 habit is required both in writing and reading the symbols, as 

 well as considerable application to become so familiar with them 

 as instantly to comprehend their meaning, especially of the more 

 comphcated formulae ; and after all, what is the great benefit 

 they are supposed to confer? A brief and easy method ofstatino- 

 the exact composition of all chemical compounds. For the 

 brevity, it is more than counterbalanced by the risk of error ; 

 for the facility, it requires considerable study to learn to do that 

 in one way which every body knows hov,' to do in another with- 

 out any study at all. But it expresses the exact composition of 

 every substance in all its minutiae — the number of atoms of bases 

 and acids, of the electro-positive and electro.negative elements — 

 and they are too complicated to be expressed in common lan- 

 guage without a tedious multiplication of words. They do 

 indeed express the exact composition assigned to the various 

 compounds by tlie hypothesis of Berzefius ; but may not all 

 those compounds be reduced to much simpler forms, and conse- 

 quently the necessity for expressing them "by this short-hand 

 character be done away v»'ith > We shall try this question most 

 fairly, by comparing the results of two or three analyses calcu- 

 lated on Berzelius's system, and on the more simple views 

 adopted in this country ; and for this purpose we may take some 

 of those copied from Beudant in the preceding pa2.es, and first 

 that of a variety of emerald (see p. 338). The "weight of an 

 atom of 



Silica = 16 



Alumina = 17* 



Glucina = 26 



Proceeding on the principles already explained, we obtain 

 the following quotients by dividing the quantity of each sub- 

 stance, as found by the analysis, by its proper atom. 



Atoms. 



^* = 429 = 8 of silica. ^ 



Id 



-— - = 105 = 2 of alumina. 



1340 1 r 1 • 



26 » 



• The weights of the atoms are from PliiUips's talile except thnt of alumina, wliich we 

 take from Ucrzehu» for the reasons given in the note (p. Mi}). 



