34 Mr. J. F. Daniell on Crystallixation. 



The tetrahedral arrangement, in fact, represents the cavi- 

 ties in the octohedral construction ; and the octohedral ar- 

 rangement would exactly fill the interstices of the tetrahedral. 



This appeal to the eye cannot be without its effect in pro- 

 ducing a conviction that such arrangements, although sym- 

 metrical, must be unstable, and that they are contrary to all 

 our ideas of the common powers of attraction in matter. It is 

 so obviously impossible, according to all our experience, that 

 solids of this kind should attract one another by their edges, 

 and not by their sides, that we are compelled to adopt the 

 unphilosophical expedient of recurring to an uncommon and 

 unknown power ; our ignorance of which will be but ill con- 

 cealed, by conferring upon it the name of polarity, or some 

 such indefinite, convenient term. 



Another observation here occurs, which I never remember 

 to have seen advanced, but which appears to me to be fatal to 

 this hypothesis. M. Haiiy, in this ambiguous choice of a 

 primitive form for fluor spar, and a vast variety of other crys- 



