Lavis — On the Structure of Rocks. 139 



together, none of them separate from each other, however long they 

 may be kept in the fluid state. Thirdly, all magmas may be 

 looked upon as originally mixtures of fused oxides, some basic and 

 other acid, it is true ; but in either extreme types there is a certain 

 amount of intermixture. We find such substances as the fats, 

 mineral oils, chloroform separating from water, or mercury from 

 either ; but we must remember that these incompatibles are built 

 up of molecules, arranged on entirely different plans, which is not 

 the case with the constituents of volcanic rocks. It may seem im- 

 probable, but I feel sure that time will show that the active cause 

 of various rock composition, at any rate, to a certain extent, will be 

 proved to result from some chemioal changes brought about between 

 an isolated portion of an original common magma and the neighbour- 

 ing rocks. Also the infiltration of saline solutions may result in 

 the bases of the contained salts, combining with the silica, and 

 liberating the original acids. The facts that support such a theory 

 are certainly few, but also those that can be urged against it are 

 equally so, and in most cases can be answered. Thus, for instance, 

 when great dykes, such as those that traverse the north of England 

 for miles, change little their composition ; and we hear, even at a 

 most recent date, such an authority as Mr. Teall arguing against 

 this theory ; it does seem in a tottering state. We must, however, 

 remember that in most cases we are only able to examine a dyke, 

 over any large area, in its horizontal extension ; but what is really 

 necessary would be to investigate such sheets of rock in their verti- 

 tical extension. There are examples in various parts of the world 

 were dykes that extend to some distance show alteration in compo- 

 sition as the rocks traversed change in character. 1 Yon Buch and 

 others have shown that in the Tyrol granite veins gradually pass 

 into basalt ones, when traversing dolomitic limestone. The basalts 

 of the Cyclopean Islands that are intrusive in a clay are most 

 markedly altered where the dykes are thinnest. It has been shown 

 that the great Whin-Sill has swallowed up beds roughly equal to 

 its own thickness. On theoretical grounds we could easily under- 

 stand an acid lava taking up limestone with its impurities, and 



1 N. S. Shaler. "Propositions concerning the Classification of Lavas, considered 

 with reference to the Circumstances of their Extrusion." — Anniversary Memoirs of the 

 Boston Society of Natural History, 1880, 



