It. A. Daly — Mechanics of Igneous Intrusion. 31 



its country-rocks. We have seen that, though the viscosity of 

 such a magma may be several hundred times that of water, the 

 roof-sections, once immersed, must sink in the magma. All 

 penologists who believe in magmatic or other differentiation 

 as operative in batholiths must i'ace the common difficulty. 



Secondly, the writer has shown reasons for believing that the 

 earth's crust at present rests on a continuous couche of basaltic 

 (gabbroid) magma, either quite fluid or ready to become fluid 

 when injected into the crust. If the average specific gravity 

 of the crust is 2*75 (a probable value), it would as a whole be 

 quite able to float on the basaltic couche, which, as noted in 

 Table II, would probably have a specific gravity over 2*90. 

 Imperfect as the numerical data are, we seem justified in con- 

 cluding that the earth's crust is now, as a whole, in stable 

 notation.* 



It may have been entirely different in pre-Keewatin (earliest 

 Archean) time when the superficial, acid couche of the primi- 

 tive earth began to solidify. Then foundering may have taken 

 place, as Kelvin imagined, and the early formed crusts could 

 have sunk a score of kilometers or more until they met the 

 denser couche below. Possibly some of the complexity of the 

 pre-Cambrian formation may be referable to this unstable con- 

 dition of the early crust. Already in Keewatin times the acid 

 shell was solidified and was then penetrated by basaltic injec- 

 tions which reached the surface, forming the heavy masses of 

 greenstones belonging to that period. Since then the crust 

 has remained essentially coherent, and through it the primary 

 basalt has, at many times and places, been erupted. It is, 

 however, quite possible that the lack of system among the 

 axes of the Laurentian batholiths and the abundance of those 

 batholiths are both explained by the thinness and weakness of 

 the crust in post-Keewatin and pre-Cambrian time. 



For Paleozoic and later batholiths there is a well-defined 

 law that they have penetrated the crust only on the sites of 

 folded geosynclinals, and that the larger batholithic axes are 

 usually arranged parallel to the respective geosynclinal and 

 mountain-range axes. 



In other words, the intrusion-history of the globe may be 

 conceived as divisible into three epochs : the first being that 

 in which the outer primary shell was becoming stable through 

 successive solidifications and founderings ; the second being 

 the post-Keewatin (Laurentian) epoch of very general inter- 

 action between the fluid basaltic substratum and acid crust, 

 without extensive founderings but with development of many 

 large, irregularly occurring batholiths ; the third, a period of 

 the localization of batholiths in certain mountain-built belts, 



*For a further discussion of this point see this Journal, xxii, p. 201, 1906. 



