236 



MINERALOGY. 



[Sect. VIII. 



when these are pure, is considered to be intunately con- 

 nected with their chemical composition, a knowledge of 

 ii becomes a useful pi-eliminary inquiry. It may be 

 almost unnecessary to remark that the specific gravity of 

 a mineral is obtained by first weighing it in air, then in 

 ^vater, that the loss it sustains in water gives the weic^ht 

 of the displaced water, and that, by dividing the weight 

 in air by the loss, or the weight of an equal bulk of water, 

 the specific grayity is obtained. Thus if W be the weight 

 of a substance in the air, W^ itn weight in water, W— W^ 

 will be that of the displaced volume of water, and 



the specific gravity. 



W 



w 



TV 



It has been above stated that with the exceptions 



Siiif 



-^^ 



arismg from dimorphism and isomorphism^ the chemical 

 composition and crystallization of mineral substances bear 

 a marked relation to each other. Whatever the forms 

 of the ultimate atoms of matter or those of the integrant 

 molecules* of the substances of the minerals may be, very 

 defi^nite and constant fundamental polyhedral solids are 

 enerally found accompanying definite chemical compo- 



ihese 



sitions (with the exceptions mentioned), and the study of 



forms has led to the science of cry stall ograpky. It 

 would be out of place here to attempt to enter upon this 

 highly interesting branch of knowledge ; it will be found 



be in powder, it is important that aD air-bubbles be remoyed, and if 

 porous, that the air in the pores be replaced by water. 



* It has been consU^red that all crystals may be derived from a 

 prism, and thus that all minerals might be composed of prismatic par- 

 ticles closely joined to each other without void spaces. On the utiier 

 hand polyhedral forms have been thought most probable, void spaces 

 being interposed between the molecules, and this the researches of M. 

 Biot on lamellur polarization are considered lo have proved. 



Vi 



tJ^' 



V* 



d 



sketc 



been 



Exp 



led 1^- 



ionif. 



t 



}ia 



upon 

 may 



inents < 



< 



rienee i 



it If 



substan 



would 1 

 genera] 



inmera 



eonditi 



circnra 



cryst 



ui 



amera 



same 



6 



each 



well 



ral 



ii 



n 



vein 



lion of 



no 



' Sim 



This ic 



'-\ti i 



1 



cl ■ 



H^, 



■'?: 



til 



