594 MANUAL OF NATURAL HISTORY. 



nature of siliceous minerals, in which the acid 

 appears to have more influence than the base in de- 

 termining their general physical properties, so that 

 in most books, even in the recent works of Dana 

 and Ansted, they are grouped together as a distinct 

 set. So long, however, as such classes are admitted, 

 the system into which they enter, though possibly 

 convenient for study or reference, must remain thus 

 far artificial. In the following pages an attempt 

 has been made to follow a classification strictly che- 

 mical, principally according to a scheme suggested 

 in the last part of Professor Dana's work, in which 

 minerals are proposed to be arranged according to 

 the elementary bodies principally entering into their 

 composition, salts being placed with their metallic 

 bases. It is perfectly impossible within such a nar- 

 row compass to do more than indicate the heads of 

 such a plan, the details requiring much working out, 

 and possibly considerable modification. In each 

 family a few of the chief genera are mentioned, as 

 illustrations ; those of more importance have their 

 more striking properties alluded to ; and, as che- 

 mistry must for the future be an indispensable in- 

 gredient in mineralogical analysis, the more simple 

 and applicable general reactions of the compounds 

 of each metal, or their action under the blow-pipe 

 are noted ; the symbols, also, of each element, as now 

 usually understood and employed, are appended for 

 reference, and to enable the reader more easily to 

 understand the formulae employed in most recent 

 treatises on this subject. 



