92 M. Keilhau's Theory of Granite, and other Rocks. 



tended to ; for, as we have recognised as a rule never to be 

 deviated from, the geologist in his postulates must always have 

 it in his eye, that, sooner or later, these must be brought to 

 accord with chemical principles ; so that it is not till, in his 

 search after a theory, he has failed in obtaining assistance from 

 his more or less perfect acquaintance with the formation of 

 crystals or analogous bodies, that he should consider himself at 

 liberty to go beyond all direct experiment. 



After these general remarks, we must proceed to inquire in 

 what other way than that proposed by the Neptunists and Vul- 

 canists, such formations could take place as may afford analo- 

 gies, or at least hints for the solution of the problem as to the 

 origin of granite, syenite, &c. 



In following up this subject, the first thing that presents it- 

 self to us is the well-known phenomena of the crystallization of 

 substances which occur in the form of vapour or gas ; but be- 

 yond this a dark region lies before us, into which, however, we 

 must penetrate. It appears that mineral bodies, or to speak 

 generally, solid unorganized bodies, are capable of changing 

 their morphological and chemical characters ; that is to say, they 

 present themselves in forms which did not exist before in their 

 place, and exhibit a combination of chemical constituents differ- 

 ent from their former one, or in short appear as new bodies in 

 both respects, without necessarily requiring that this change 

 should be preceded by their having been in a fluid state. I say 

 it appears that this is the case, for that a movement of the whole, 

 or at any rate of part of the substance of which the bodies pro- 

 duced in this manner consist, and in the case of newly formed 

 crystals at least, a very free movement must be supposed to have 

 previously taken place ; and the possibility of this we can only 

 conceive in a state of fluidity. It is nevertheless a fact that 

 formations take place partly with a material, which, to our ob- 

 servation, was, and continues to be, what we call firm and rigid, 

 that is, neither in a liquid nor in an elastic state of fluidity ; 

 and partly in a solid medium, which, according to our common 

 ideas, seems not capable of allowing any movement of the sub- 

 stances in it. It is extremely unfortunate for geology that the 

 whole of this field for important investigation has been left al- 

 most quite uncultivated. Thus there exists a metallurgical pro- 

 cess which exhibits in a striking manner the fact we have men- 



