286 Formation of Compound or Twin Crystals. 



But if it is possible to determine the form of a molecule in one in- 

 stance from a crystal, why not, it may be asked, in all? A deduc- 

 tion from the Cube of Copper that its molecule was a sphere, would 

 lead to the same inference from the Cube of Pyrites, or that of 

 Common Salt, which are compound bodies- This method, and it 

 would seem to be the only possible one, therefore, proceeds on the 

 incorrectness of this theory. 



It may be assumed (with what propriety, it is difficult to say,) 

 that the molecules of the elements are equal spheres. . In doing so, 

 however, the inconsistency between the supposition and the crystal- 

 line forms of Sulphur, Selenium, Tellurium, Iridium, Arsenic, Anti- 

 mony, &c. is forgotten. If the assumption is correct, the unavoid- 

 able conclusion must be drawn, that there is no similarity in any in- 

 stance, between the primary form of a mineral and its molecule ; for 

 that relation cannot be supposed to exist elsewhere if not in the 

 above substances. But this resulting principle is too improbable to 

 be earnestly brought forward. The invariable proportional dimen- 

 sions and angles of the primary forms, the symmetrical arrangement 

 of secondary planes, as also the optical properties of minerals, are op- 

 posed to such an hypothesis. 



The mind may indeed possess a vague indefinite idea of the ag- 

 gregation of particles by some law acting independently of their 

 forms ; but it is a peculiarity of this idea, the greater the effort to 

 bring it distinctly before the mind and comprehend it, the more in- 

 comprehensible it becomes. 



M. Gaudin states that in the construction of the molecule of Feld- 

 spar, he was guided by the rules of " equilibrium and symmetry" 

 These rules however could have been derived from no other source 

 but the unassisted judgment, whereas in a question of this kind, 

 they ought to have proceeded from some fixed laws, a consequence 

 of causes analogous to, if not identical with those known to exist. 

 The following are his three classes of crystals in his own words : 

 **■ 1. Ceux ou les files d' atomes n' existent pas plus que dans les li- 

 quides transparens, mais ou les grandes espaces vides sont reguliere- 

 ment repartis (crystals of the Tesseral system) ; 2. Ceux ou les files 



atomes ont une direction unique (eristaux a une axe) ; 3. Ceux 

 ou les files d' atomes affectent plusieurs directions, (eristaux a deux 

 axe.") 



The formation of a molecule of a compound resembling the pri- 

 mary form, requires a separation of the atoms of the constituent parts 



