Theory of Cryftallifatibn. 163 



6f the cryftal. Thas in the laminae applied on ABCD 

 (Jig. 29), all the edges ahfwering to CD would be on a le- 

 vel with CDFGj of which they would form' a prolonga- 

 tion ; and in the laminae applied on D C G F all the edges 

 analogous to the fame ridge C D would be on a level with 

 ABCD, from which there would neceffarily refult a re-en- 

 tering angle oppofite to the falient angle formed by the two 

 faces ABCD and CDF G. But the re-entering angles 

 appear to be excluded by the laws which determine the 

 formation of fimple cryftals. The folid will increafe then 

 in thofe parts to which the decrement does not extend. 

 But as this decrement is alone fufficient to determine the 

 form of the fecondary cryftal, we may fet afide all the other 

 variations which intervene only in a fubfidia'ry manner, ex- 

 cept when it is wifhed, as in the prefent cafe, to conftruct, 

 artificially, a folid reprefentation of a cryftal, and to exhibit 

 all the details which relate to its ftructure. 



The fuperior face of the fecond lamina will be like 

 A', G', L', K', (Jig. 32), and this lamina muft be pfacrfi 

 above the preceding in fuch a manner that the points a" t b 1 ', 

 c", d", may anfwer to the points a' 3 V , c', J',- [Jig- 3 1 )' 

 which will leave uncovered the fquares that have their 

 exterior angles fituated in Q, S, E, O, V, Ty M> G, &c. 

 and continue to effect the decrement by one range. It is 

 here feen that the folid increafes fucceffively towards the 

 analogous edges at A B, B C^ C D,- A D (Jig. 30), fince be- 

 tween A' and L/, for example, (Jig. 32), there are thirteen 

 fquares, whereas there are only eleven between Q V and L I 

 (fig' 3 1 )- B ut as the effect of the decrement contracts 

 more and more the fuvface of the laminae in the direction of 

 the diagonals, nothing is ncceffary but to add towards 

 the edges that do not decreafe, a fingle cube denoted by 

 A', G', L or K', (Jig. 32), inftead of the five which termi- 

 nate the preceding lamina along the lines QV, GF, LI,- 

 ON, (Jig. 31). 



The large faces of the laminae of fuperpofition, which hi- 

 therto were Octagons, Q V G F I LN O, (Jig. 31), having 

 JVI 7> arrived 



