L] THEORY OF IGNEOUS ROCKS AND VOLCANOES. 7 



vey of Canada for 1856, p. 479, that the reaction between 

 alkaline silicates and carbonates of lime, magnesia and iron at a 

 temperature of 100 C. gives rise to silicates of these bases, and 

 enables us to explain their production from a mixture of car- 

 bonates and quartz, in the presence of a solution of alkaline 

 carbonate. I there also suggested that the silicates of alumina 

 in sedimentary rocks may combine with alkaline silicates to 

 form feldspars and mica, and that it would be possible to crys- 

 tallize these minerals from hot alkaline solutions in sealed 

 tubes. In this way I explained the occurrence of these sili- 

 cates in altered fossiliferous strata. My conjectures are now 

 confirmed by the experiments of Daubree, which serve to 

 complete the demonstration of my theory of the normal meta- 

 morphism of sedimentary rocks by the interposition of heated 

 alkaline solutions. 



But to return to the question of intrusive rocks : Calculations 

 based on the increasing temperature of the earth's crust as we 

 descend, lead to the belief that at a depth of about twenty-five 

 miles the heat must be sufficient for the igneous fusion of ba- 

 salt. The recent observations of Hopkins, however, show that 

 the melting points of various bodies, such as wax, sulphur and 

 resin, are greatly and progressively raised by pressure, so that 

 from analogy we may conclude that the interior portions of the 

 earth are, although ignited, solid from great pressure. This 

 conclusion accords with the mathematical deductions of Mr. 

 Hopkins, who, from the precession of the equinoxes, calculates 

 the solid crust of the earth to have a thickness of 800 or 1,000 

 miles. Similar investigations by Mr. Hennessey, however, as- 

 sign 600 miles as the maximum thickness of the crust. The 

 region of liquid fire being thus removed so far from the earth's 

 surface, Mr. Hopkins suggests the existence of lakes or limited 

 basins of molten matter, which serve to feed the volcanoes. 



Now the supposed mode of formation of the primitive molten 

 crust of the earth would naturally exclude all combined or 

 intermingled water ; while all the sedimentary rocks are neces- 

 sarily permeated by this liquid, and consequently in a condition 

 to be rendered semi-fluid by the application of heat as supposed 



