top On Crystallography, 



1. Acl<liferous substances ; composed of an aciJ united 

 to an earth or an alkali, and sometimes to both. 



2. Earthy substances ; into the composition of which 

 earths only enter, united sometimes with an alkali. 



3. Combustible substances (not metallic). 



4. Metallic substances. , 



We have also explained the reasons which induced us 

 to choose the base rather th^n tht.* acid for characterizing- 

 the genera of the lirst cla^ss, and those to which metallic 

 substances belong, which have also an acid for one of their 

 component principles. Hence, in the last place, the ne-» 

 cessity of modifying, by the help of a very simple inversion, 

 the specific names adopted by modern chemists, when they 

 say sulphated lime instead of sulphate of lime ; carhuretted 

 iron, instead oi carburet of iro?iy &c., in order to modify 

 the nomenclature to the system itself, and to preserve that 

 precision and regularity so happily introduced by Linnaeus 

 into the language of natural history. 



The care which we have taken to restrict our method to 

 what is founded on exact and precise experience, dictated 

 that we should exclnde certain substances, the nature of 

 ■which is not yet sufficiently verified to admit of our de- 

 ciding whether they constitute distinct species, or if they 

 may ue referred to any of the species already classified. 

 We shall arrange- these substanceis in a first appendix, in 

 which we shall give the description of them, detailing the 

 gurmisesw^ich their characters may have already originated, 

 as to what they would become if better known. A second 

 appendix will contain the substances which are only ad- 

 mixtures of different species, and among which are found 

 the aggregates known by the name of rocks, and which 

 we shall therein more particularly describe. Lastly ; there 

 will be a third appendix for the productions of volcanos, 

 and for those of subterranean fires, but not volcanic. 



After all it must be remembered, that our methods, even 

 when best managed, represent nature but imperfectly ; and 

 after having conducted us a certain length they leave us 

 to ourselves. It should seem as if they had only been 

 composed from selected pieces, in which tlie characters 

 which they indicated had a clearer or better pronounced ex^ 

 pression. Incomplete as they are, however, they have the 

 precious advantage of putting our ideas in order, and pre- 

 pare us for a more detailed study of their object. The in- 

 struction having been once begun by the exact application 

 of the principles to selected specimens, is more easily 

 and more happily terminaiL^d> by the close observation of 



every 



