254 AN INQUIRY INTO THE [Ch. XII. 



these rocks under circumstances favourable to the exertion of 

 the attraction of cohesion, effect the same object by different 

 means. 



This structure of rocks appears, then, to be a species of 

 Crystallisation : this idea has been already advanced by Pro- 

 fessor Jameson and other geologists, but has been generally 

 discountenanced, as a vague and unscientific proposition. 

 But in what respects does it differ from crystallisation ? In 

 crystals, as in rocks, the integrant particles are combined and 

 arranged by the attraction of cohesion into forms more or 

 less geometrical. It is true, that rock concretions do not 

 exhibit such symmetrical figures as the most perfect crystals ; 

 but this may in some measure be accounted for by the more 

 complicated composition of these concretions : so that their 

 forms are not the simple result of the aggregation of identical 

 particles, but the balance of diverse powers, each tending to 

 produce a different form. And it is also true, that we some- 

 times cannot detect any regular forms, though the mass is 

 evidently concretionary, it being intersected in every direc- 

 tion by smooth planes, which give rise to irregular angular 

 pieces : yet, even in this case, we find an analogy among 

 crystalline substances, whether simple as in salts, or compound 

 as in the heterogeneous crystalline masses occurring in the 

 cavities of veins ; in all such cases, in short, as are attributed 

 to a confused crystallisation ; and this is no uncommon circum- 

 stance, for we find a hundred instances in which a substance 

 is crystallised in an indeterminate mass, to one in which it is 

 perfectly and regularly formed. 



On this subject, however, it must be confessed, that we do 

 not possess sufficient information to draw any positive infer- 

 ences : but the case is strong enough to demand more atten- 

 tion than it has hitherto received. It opens an additional 

 and almost unexplored region, which ought to be carefully 

 examined, for it cannot fail to enlarge our knowledge of the 

 earth's structure. 



It would occupy too much space to particularise all the 

 points that are elucidated by referring the lines or joints of 



