VII.] 



CRYSTALLIZATION. 



83 



All crystals are of the same chemical constitution 

 throughout: differing in this from organisms, in all of 

 which, except the lowest, there is a difference of com- 

 position in the different parts. But in some crystals the 

 structure is not the same throughout. Sir David Brewster 

 has described^ specimens of Apophyllite, a square pris- Tesselatcd 

 matic crystal, consisting of crystals symmetrically united iite°^ '^ " 

 together, some of wliich, when examined by polarised 

 light, prove to be uniaxial, and some biaxial : the optical 

 character of each single crystal being in some way deter- 

 mined by its position in the compound one. One of the 

 simpler forms described has a horizontal 

 section like this : the central square 

 prism being uniaxial, and the surround- 

 ing irregular ones biaxial. No differ- 

 ence of structure is discernible by any 

 other means, and no joinings between 

 the prisms are to be perceived. 



Different parts of a crystal of Analcime have different Analcime. 

 optical structure. Analcime is a cubic crystal; but, un- 

 like other cubic crystals, it has a very decided polarising 

 structure : and, unlike all other known crystals whatever, 

 it consists of parts that have different optical properties, 

 not sharply separated as in the tesselated Apophyllites 

 described above, but graduating into each other. If the 

 cube of Analcime be dissected by planes passing through 

 the twelve diagonals of its six faces, these planes will be 

 found to be places of no polarisation. Every other part 

 of the crystal polarises with a force proportional to the 

 square of its distance from the nearest of those planes.^ 



All perfect crystals have cleavage planes : that is to Cleavage, 

 say, directions in which they split more easily than in 

 other directions. This property of course depends on 

 molecular structure. Cleavage planes are always parallel 

 to faces which are possible ones in the species, and con- 

 sequently faces artificially or accidentally produced by 

 cleavage belong to the crystalline species as truly as those 

 with which the crystal was originally formed. 



^ Edinb. Phil. Journ. vol. i. 



* Pereira on Polarised Light, p. 194. 



g2 



