436 SCIENCE PROGRESS. 



than 6 and atomic volume less than 6. Each species 

 receives a Latin name ; e.g., among the Garnets are 

 Granatus Almandicus, Granatus Manganosus, etc. 



The chemical relations of the various species are further 

 supposed to be similar to those of the carbon compounds ; 

 what is commonly known as an isomorphous group of 

 minerals is therefore to be regarded as a homologous series 

 differing by multiples of a certain radicle ; if in two or more 

 minerals having a similar empirical composition the value 

 of v is the same they may be regarded as having a 

 metameric relationship ; if the value of v differs they are 

 isomeric homologues which are differently polymerised. 

 In a previous essay it had, for example, been suggested 

 that Calcite has the formula C 30 Ca 30 O go and Calamine 



**-40 ^^40 ^40' 



This is not the place to enter further into these 

 questions or to discuss the new mode of chemical notation 

 introduced by Sterry Hunt in conformity with his views ; 

 our present purpose is to compare this and the other 

 natural systems with the orthodox classification of the 

 present day. 



It may be noted, however, in passing that the principle 

 of polymerisation is introduced in a still more comprehen- 

 sive manner by Goldschmidt, who, in 1890, proposed to 

 consider all silicates as polymers of Si 2 in which a portion 

 of the silicon is replaced by other elements whose com- 

 bined valency is the same. In this scheme all silicates 

 belong to the type n Si 2 + p H where H is water or 

 such a radicle as KF ; thus Orthoclase is (Si 2 ) + in which 

 one atom of Si is replaced by the radicle K 2 Al ; the 

 Garnet Ca 3 Al 2 Si 3 O i2 is 2 (Si 2 ) 3 in which one atom of 

 Si is replaced by Ca 2 and two atoms of Si by Ca Al. 



Other systems of mineral classification, such as that 

 adopted by De Lapparent in his admirable Coztrs de 

 Mine'ralogie, which arrange the species not by their 

 characters but by their origin and mode of occurrence in 

 nature, we may pass by as not scientific systems, whatever 

 may be their educational value or convenience ; since, if 

 anything is certain, it is that the essential characters and 



