On Crystallography. !04 



fi'C shall presently speak of, may lead to the determinatioa 

 of their primitive forms, at least with a great probability of 

 accuracy. 



Having adopted the word primitive form, in order to de- 

 signate the nucleus of crvstals, we shall call secondanj forms 

 such varieties as differ from the primitive form. 



In certain species, crvstallizatum also produces this last form 

 immediately. There exist, for example, calcareous crvstals 

 which differ in no respect trom the rhomboid which we extract 

 from the regular hexahedral prism, and from the other varie- 

 ties which we have mentioned. Besides, it frequently happens 

 also, that among the faces of a secondary crystaj there arc 

 some which are parallel to those of the primitive form. Thus 

 we find crystals of carbonated lime, which are similar to 

 that of fig. 3, and on which crystallization has left pentagonal 

 planes, such as A Eoo r, AE/i/iG, &c., situated like those 

 which we expose by dividing the hexahedral prism repre- 

 sented in fie:. 1 . In such cases as these, the route is as it were 

 traced out before hand, previous to arriving at the nucleus. 



We mav define the primitive furm a solid of a constant 

 form, engaged svmmetiii-allv in all the crystals of one and the 

 same species, and the faces of which follow the directions of 

 the ki.ninae which form these crystals. 



The primitive forms hitherto observed are reduced to six, 

 viz. : the paralltlopipedon, the octahedron, the tetrahedron, 

 the regular hexahedral prism, the dodecahedron with rhom- 

 bous planes, all equal and similar, and the dodecahedron with 

 triangular planes, composed of two straight pyramids joined 

 base to base. 



Forms oj integrant Mdltmles. — The nucleus of a crystal 

 is not the last term of its mechanical division. It may al- 

 ways be subdivided parallel to its different faces, and some- 

 times in other directions also. The whole of the surround- 

 ing substance is capable of being divided by strokes parallel 

 to those which lake place with respect to the prinntive form. 

 Reasoning here as we have done with respect to the me- 

 chanical division of common salt, we conclude that the 

 limit of the mechanical division which we may operate in 

 any given crystal, ought to give the form of the integrant 

 molecule peculiar to the kind of mineral of which this cry- 

 stal is the primitive. 



II the' nucleus be a parallelopipcdon which cannot be sub- 

 divided except by blows parallel U> its faces, like that which 

 Jakes place with respect to carbonated lime, it is evident that 

 the integrant molecule will be similar to this nucleus itself. 

 3ut »t may happen that the parallelopipedoa admits of 



further 



