86 GEOLOGY OP THE HIGH PLATEAUS. 



stitution of the numerous kinds and to their degrees of affinity; second, 

 those having reference to their genesis. In other words, we desire a 

 formula which shall express what the rocks are and the causes which made 

 them what they are. It may be said at once that we liave no knowledge 

 of the genesis of volcanic rocks sufficient to make a coherent formula, or 

 out of which we can construct a system of causation, however crude. We 

 know that they came up out of the earth in a molten condition, and that 

 is all we can confidently say of their origin. Our classification, therefore, 

 must, from the necessities of the case, be confined to an expression of what 

 we know concerning their physical constitution. In this direction our 

 knowledge is sufficient to justify an attempt to formulate it. 



Let us look first at those physical properties which are common to all 

 volcanic rocks, and which, therefore, serve to distinguish them as a cate- 

 gory from all other categories ; if, indeed, such a distinction really exists. 



1. All volcanic rocks have been in a state of fusion at a high tem- 

 perature. • 



2. All volcanic rocks have been displaced from unknown depths in 

 the earth, and have risen in a fiery, liquid condition, either to the surface, 

 where the}^ have outflowed as lavas, or have intruded themselves, part-way 

 up, among colder overlying rocks, where they have quietly solidified. 



3. They consist of aluminous silicate, combined with lime, magnesia, 

 soda, and potash; iron is very rarely absent — perhaps never wholly want- 

 ing. Moreover, the quantities of these several oxides, though varying, 

 have tolerably narrow ranges of variation. Thus the silica never materi- 

 ally exceeds 80 per cent, nor falls sensibly below 45 per cent. ; the alumina 

 ranges from 10 to 20 per cent., the lime fi-om 1 to 10 per cent, &c. 



4. All volcanic rocks consist of an amorphous base, holding crystals, 

 except, however, some intrusive rocks, which appear to be wholly crystal- 

 line. In some obsidians, on the other hand, crystals are exceeding rare, 

 though probably no great mass of obsidian is wholly without them. 



Although it seems as if there ought never to be any difficulty in dis- 

 tinguishing a volcanic rock from any belonging to other categories, yet 

 this difficulty sometimes arises. A rock may have been fused and dis- 

 placed from its seat; it may have the chemical constitution and "half- 



