Formation of Com/pound or Twin Crystals. 289 



glesj are North, and those about the other, South. Thus In Crys- 

 tals of Tourmahne there is no opposition between the Crystalloge- 

 nic poles and the electric induced by heat. Probably also in the 

 oblique prisms, the poles about a dominant solid angle are of the 

 same kind. It may be also inferred that the poles about an acute 

 edge in the right prisms are of the same kind, as marked in Fig. 5. 

 Farther than this, it is at present impossible to distinguish the poles 

 of the axes in the different Primary forms. 



The molecules of crystals, governed by the usual principles of 

 attraction, the repulsion of like poles and the attraction of unliTce, 

 will assume the arrangement given in Fig. 9. The general action 

 of the poles on one another, will cause the axes of the molecules to 

 assume a parallel position, and also a uniformity in the direction of 

 similar poles. Such is the general principle in the architecture of 

 the Crystalline Solids. We may now notice the apparent excep- 

 tions to it, exceptions which are a consequence of the general prin- 

 ciple, and of which it may be correctly said, " exceptio probat re- 

 gulam." 



Two molecules assuming simultaneously their axes of attraction, 

 would have any situation towards one another, were there no mutu- 

 al influence between them. But since, from the very nature of at- 

 traction, they must necessarily influence one another, guided by this 

 attraction, they will always assume the position in Fig. 9, unless 

 they are nearly or entirely inverted as in Fig. 10. In this case the 

 strong attraction between the adjacent north and south poles, and 

 the not unfavorable position of the other axes to the occurrence, 

 may counterbalance the tendency of the molecules to invert them- 

 selves in order that the joining axes may be in the same straight 

 line, and hence may bring them together as in Fig. 11, a posi- 

 tion nearly as natural as that of Fig. 9. Consequently a com- 

 pound nucleus is formed, each half of which now commences to 

 act independently of the other, although in connection with it, and 

 the issue is a compound or twin crystal. Such may be considered 

 the origin of those compound crystals, whose composition has taken 

 place parallel to a primary face. That the accident to which they 

 owe their origin should have happened, results as has been said, 

 from the nature of the molecules, and the non-occurrence of it, might 

 have been adduced as a strong argument against the whole theory. 



This species of composition can only be detected among those 

 crystals which have at least one obhque interfacial angle, unless it 



Vol. XXX.— No. 2. 37 



