l^etween the Earthy Carbonates and Alkaline Nitrates. 48 1 



to the analogy in atomic constitution generally observed 

 among compounds known to assume the same form? 



Between the formula Ca C and K N there is no obvious 

 analogy. If we represent the electro-positive elements in each 



by R, the formulae become R and R, which also belong to 

 very different groups of compounds, analogous to the sesqui- 



oxide and first acid of manganese Mn and M. But if we 

 halve the equivalent of potassium (for the reasons given in a 

 former number of this Journal) and adopt Berzelius^s weight 



for nitrogen, the formula for nitre becomes K N, or taking the 



positive elements together R4 Og or R the same as the formula 

 for the dimorphous carbonates. 



This mode of establishing an analogy in atomic constitution 

 between these isodimorphous compounds is very short and 

 simple, but there are very strong reasons why we ought not 

 at present hastily to adopt it. In all compounds known to 

 crystallize in identical forms an analogy in atomic constitution 

 has been observed to prevail, with few exceptions, not only 

 between the crystalline compounds as a whole, but between 

 the several compounds of two or more elements, by the union 

 of which they are supposed to be directly formed. Thus, 



Ag S, and Na Se, which crystallize in forms nearly identical, 

 are analogous not only in containing each two equivalents of 



positive to four of negative elements (R), but also in containing 



an analogous oxide of a metal (R) united to an analogous acid 



(R). The salts of ammonia present an exception to the first 

 part of this rule, the sulphate or seleniate of this base being 



represented by a formula in which the oxide R or R 

 silver and soda salts is replaced by N H4 O. StillTh^ ^"^" 

 logy holds in regard to the acid, and we draw a doub^^ i*^" 

 f^rence in both cases, that the entire salts, namely, are iso- 

 morphous, and that the acids and bases also are isomorphous 

 each with each, and may mutually replace each other. 



The titaniate of the protoxide of iron Fe Ti, in replacing 



the peroxide of iron Fe in Ilmenite and some allied minerals, 

 presents an example which at first sight would appear to justify 

 the method above suggested for establishing an analogy be- 

 tween the nitrate of potash and the carbonate of lime. But 

 Phil. Mag. S. 3. Vol. 12. No. 77. June 1838. 2 S 



