352 Mr. Children s Summary View of [May, 



ciples is often replaced by one or more principles of the same 

 formula, whose quantity varies indefinitely in diti'erent analyses, 

 but is always such that its oxygen is equal to that of the princi- 

 ple replaced : hence if we take on the one hand the oxygen 

 contained in the common acid, and on the other the sum of the 

 quantities of oxygen contained in the bases, we obtain numbers 

 which are precisely in the same ratio to each other that they 

 would be if the compound were perfectly pure. We are even 

 led to the- knowledge of cases in which one of the immediate 

 principles is replaced by another with a totally different acid, 

 the base either remaining the same, or being itself different. 

 Thus silicates of lime are replaced by aluminates of the same 

 base, or by aluminates of bioxide of iron, &c."* 



M. Beudant then goes on to illustrate the preceding observa- 

 tions by examples, and gives a sort of receipt for making garnets, 

 or rather endeavours to show how one compounded of many 

 elements may be divided into several others of more simple 

 composition. 



We shall quote, with some abridgment, his first example. 



" There are garnets obviously of the following composition : 



Silica 41 containing oxygen 20-60 or 2 atoms 



Alumina 22 10-27 1 



Lime 37 10-39 1 



Too 



" which indicates 2 atoms of silicate of alumina, plus 1 atom of 

 sihcate of lime ; and a series of direct analyses presents a multi- 

 tude of other results that can only be explained by calculating 

 them on the atomic system. Thus the connexion between the 

 following analysis and the preceding is by no means obvious. 



SiHca 37-00 containing oxygen 18*61 



Alumina 13-50 6-30 



Lime 29-00 8-14 



Magnesia 6-50 2-51 



Trioxide of iron 7-50 2-30 



Trioxide of manganese 4-75 1-41 



98-25 



" If we collect the oxygen of the bases of the same order, 

 namely, the alumina, trioxide of iron, and trioxide of manganese, 

 on the one hand, and that of the lime and magnesia on the other, 

 we find that the quantities of oxygen in the acid and bases are 

 not far from the ratio of 2, 1 and 1, consequently the new gar- 

 net very nearly harmonises witii the former. 



* Traite Elementaire, &c. p. 248, 



