1 826.] affecting Chemko- Miner alogical Si/stems. 383 



resemble the relation of genus and species in zoological classifi- 

 cation. The genus is determined by the chemical formula and 

 the geometrical form ; the species by the elements. For greater 

 clearness, let us take an example from the garnet. Its crystal- 

 line form is generally known, and the formula of its compositionj 

 according to M. TroUe-Wachtmeister (R signifying radical), is 



R' Si + 2 R Si. These two formulae determine the genus 



garnet. M. Wachmeister has proved besides, that R may be 

 lime, magnesia, protoxide of iron, or manganese, either one of 

 them alone, or several, or even all of them together ; and that 



R may be either alumina, or deutoxide of iron, sometimes alone, 

 sometimes combined with each other. Not less, therefore, than 

 eight species, or prototypes of different garnets, must result 

 from these principles, and their mixture produces such numerous 

 varieties that it would be in vain to attempt to distinguish them. 

 I shall give a second example from another species of mineral, in 

 which isomorphism is less frequently met with. According to 

 analysis, chabasie is composed of CS^ + 3 AS- + 6 Aq, and a 

 small portion of lime which they contain is represented by pot- 

 ash. I have lately analyzed a chabasie, given me under the 

 name of Levyine, in which a small part of th^ilime was replaced 

 by potash and soda. M. Arfwedson analyzed a Scotch chabasie, 

 in which almost all the lime was replaced by soda and potash. 

 It is clear, therefore, that some chabasies chiefly contain lime, 

 and others soda ; that in all, the bases lime, soda, and potash, 

 may replace one another in variable proportions, and thus 

 chabasies from different localities may be differently composed ; 

 but nevertheless ihey retain the same general formulas of com- 

 position. According to M. Beudant, crystals belonging to the 

 rhomboidal system have, in their isomorphous substitutions, 

 angles similar but not absolutely identical (whilst in the regular 

 system it is always the reverse) ; so that, after having accurately 

 measured the angles of a dolomite, we can determine the relative 

 quantity of lime and of magnesia, in the carbonate of lime and 

 of magnesia, from the measurement of their angles taken sepa- 

 rately, which are very similar, but not perfectly equal. If such 

 be also the case with the bisilicate of soda, lime, and potash, in 

 the rhomboidal crystals of chabasie, it is clear that mineralogists, 

 accustomed to measure crystals accurately, will find chabasies 

 with different angles ; and it would be as absurd to make dis- 

 tinct species of them, as of dolomites composed of lime and 

 magnesia in different proportions. I presume that the new 

 name of Levyine, given to the chabasie which I examined, was 

 derived from a similar circumstance. Here then we have genera, 

 Bpecies, and varieties, or if the word genera be objected to, we 



