440 JOSEPH BARRELL 



His theory is constructed however for action within the roof of a 

 magmatic substratum. Daly postulates a zone of tension, but the 

 mechanism suggested by Lane does not require this and would seem 

 to apply better to the region of generation of magmas, for there the 

 cubic compression is enormous, it is not a zone which has been 

 subjected to cooling, and therefore it is difficult to conceive of the 

 cause of a system of tensile stresses within the asthenosphere. 



We are now prepared to draw a closer analogy between the 

 physical conditions of the asthenosphere and those of a glacier, 

 noting the likenesses and also the unlikenesses. In the summer, 

 in the case of Alpine glaciers, heat is supplied to the surface of the 

 glacier until it is warmed to the melting-point, and part of the ice 

 absorbs the amount required by the change of state and passes into 

 water. This trickles along the surface until a fissure is met and 

 the water sinks by force of gravity toward the bottom. Near 

 the snout of the glacier the temperature of this deeper part may 

 thus be more dependent upon this convection than upon direct 

 conduction through the ice. The winter freezing tends to chill 

 the deeper ice and slow its motion, but during the summer the 

 descending water tends to raise the temperature toward the freezing- 

 point. In parts of the glacier where the heating is more effective 

 than the cooling, the waters drill channels and gather at the base 

 of the glacier into streams, reaching finally the outer world. The 

 descent, and gathering, and englacial flow of glacial waters is 

 analogous to the rise and convergence of streamlets of molten rock. 



In order to account for lateral mass movement within the 

 asthenosphere, an imperfection of isostasy between continental 

 interior and ocean basin, giving an isostatic gradient or slope toward 

 the continental interior, seems a necessary postulate. Such a 

 gradient is so low that it has not yet been sifted from those irregu- 

 larities of mass which are owing to the strength of the outer crust, 

 or, it may be, obscured by great compressive bowings of the crust. 

 This isostatic gradient, the slope required to generate movement 

 within the asthenosphere, is far lower than that of the surface of a 

 continental ice sheet. The failures of the analogy between the 

 asthenospheric and glacial states are then as instructive as the 

 agreements. The glacier is thin and broad. Friction on the bot- 



