Prof. Naumann on Mineralogical Classification. 39 



together in three very natural orders, viz., glances, pyrites, 

 and blendes. 



In regard to all combinations partaking of the nature of salts 

 (with the exception of what are termed sulphur-salts, and 

 the like), the distinction of heavy and light metals and 

 their oxides seems to be of the highest importance, as has been 

 acknowledged by Hauy and Beudant, and more prominently 

 brought forward by Gmelin. We must, more particularly, 

 make this distinction available in regard to the silicates and 

 the other salts containing oxygen, in so far as we must divide 

 these, according to this difference of their bases, into metallic 

 and non-metallic silicates and salts. Double salts, with bases 

 of both kinds, or those salts in which the oxides of the light 

 metals are frequently, and in a great measure, replaced by 

 oxides of the heavy metals, may be considered as amphoteric 

 salts. 



The silicates, and the aluminates which approach them so 

 nearly, are distinguished in so remarkable a manner from all 

 the other combinations of the mineral kingdom partaking of 

 the nature of salts, that they must necessarily be arranged 

 in a separate group. 



The distinction of the hgdrous and anhydrous condition, 

 must be pai'ticularly attended to in classification, not only 

 because it can be ascei'tained by a very simple experiment, 

 but also, because, in most cases, it is intimately connected 

 with the whole habitus, and with the mode of formation of 

 the mineral. 



Guided by these and similar principles, I submit to the in- 

 dulgent criticism of mineralogists and chemists, the follow- 

 ing attempt to arrange the mineral kingdom in series. As this 

 arrangement is neither based solely on chemical, nor solely 

 on physical, properties, but on both kinds of properties taken 

 together, it must naturally happen, that sometimes the one 

 set of properties, and sometimes the other, must receive a 

 predominant degree of attention, according as the physical 

 habitus or the chemical constitution may seem to possess 

 greater weight in the determination. 



The importance assigned to solubility in water, by found- 

 ing on that character alone two separate classes, will not be 



