lOS On Crystallvgraphy . 



further sections in other directions than the former. Let us 

 conceive for example, that it is a rhomboid A A' K H (fig. 10), 

 directly divisible parallel to the six rhombuses which ter- 

 minate it, and with the help of planes, each of which passes 

 bv an oblique diagonal A O, by the axis A' A, and by the 

 edoe A' O, comprehended between the same diagonal and 

 -the axis. These sections will detach six tetrahedrons, which 

 have been figured separately around the rhomboid, in posi- 

 tions analogous to those which they had when joined in one 

 •single body, in suoh a way that we follow as it were with 

 the eye the species of decomposition of the rhomboid from 

 which they proceed. Now these tetrahedrons represent the in- 

 tegrant molecules of the substance of which the rhomboid is 

 the primitive form. Such is the structure of the tourmaline. 



Let u3 take another substance, such as phosphated lime, 

 the primitive form of which is the regular hcxahcdral prism. 

 In this case the molecule will be still different from the nu- 

 cleus, although this last cannot be subdivided except 

 parallel with its faces, i. e. with its two bases and its six 

 panes. This subdivision will lead us to triangular prisms, 

 the assemblage of which composes the entire prism, as mav 

 easily be perceived by inspecting fig. 40, PI. V; we there see 

 one of the bases of the prism divided into equilateral trian- 

 gles, each of which is the base of a small triangular prism 

 which represents the integrant molecule. 



Now we shall presently find that we may reduce the forms 

 of the integrant molecules of all crystals to the three pre- 

 ceding forms, which are the tetrahedron, or the simplest of 

 the pyramids : the triangidar prism, or the simplest of all 

 the prisms; and the paralltlopipedon, or the simplest among 

 the solids, which have their faces' parallel two and two. 

 And since four planes at least are necessary for circum- 

 scribing a space, it is evident that the three forms in ques- 

 tion, in which the number of faces is successively four, five, 

 and six, have still, in this respect, the greatest possible sim- 

 plicity. If these forms, I repeat, are not those of the true in- 

 tegrant molecules employed by Nature, they deserve at least 

 to supply their place in our conceptions, the more especially 

 as it is with but very scanty materials that we succeed in 

 establishing a theory which embraces so many various results. 



Several naturalists have thought that the integrant mole- 

 cules of crystals were simple laminas, whose thickness was 

 incomparably less than their other dimensions, and not 

 small solids, the thickness of which was equal, or at least 

 in proportion to their breadth and length. In the Journal 

 d£S Mines, No. 28, p. 305, I have detailed the numerous 



and 



