84 GEOLOGY. 



Difficulties Encountered by the Gaseo-Molten Hypothesis. 



Quite apart from the question of its general truthfulness, two 

 sources of serious dissatisfaction with the preceding view of the early 

 earth-stages have grown out of recent investigations, the one arising 

 from failure to find any great basal formation bearing the distinctive 

 characters of an original crust; and the other, from adverse evidence 

 relative to the prodigious atmosphere postulated. 



1. Evidence Relative to an Original Crust. 



The theory of a molten earth carries the presumption that the 

 liquid mass arranged itself in concentric layers according to the spe- 

 cific gravities of the constituents, if convection was not so vigorous as 

 to prevent this. As the granitoids constitute the lightest class of 

 igneous rocks, their molten magma has been, very consistently, assigned 

 to the outer zone of the molten earth, while the magmas of the heavier 

 neutral and basic rocks have been assigned to lower layers. This 

 conception was long ago set forth definitely by Durocher, who postu- 

 lated an outer zone of rock derived from an "acid magma", and a 

 lower zone derived from a "basic magma." 1 ^ 



If, however, it be assumed that convection was too effective to per- 

 mit such concentric arrangement of liquid layers, the [Conception of 

 an intimate mixture of material must be substituted for that of differ- 

 entiated layers, but the odter 2fone of rock must probably still have 

 been homogeneous in the larger sense, for the conception of "perma- 

 nent areas of lighter rock in one region, and heavier rock in another, 

 at the same time and in the same horizon, is in^nsi§Jrt ; Fith the 

 principles of hydrostatic equilibrium. M \ > i > l 



The theoretical crust a universal identifiable substratum.— In either 

 case^a characteristic surface of lava, either sensibly homogeneous 

 or homogeneously mixed, and in approximate hydrostatic. equilibrium, 

 should have constituted the material which solidified into the original 

 crus^.,, ^TJfe i ;;Cru,s,t should, therefore, be a universal substratum of 

 a distinctive character and susceptible of definite identification. A 

 specially 1 complete state of crystallization, of the very 'coarse type, 



1 A clear statement of this logical view is given in Haughton's Manual of Geology, 

 p. 3, to which is added a translation of Durocher's notable essay on comparative 

 petrology, in which the doctrine is elaborated. 



