914 HARRY FIELDING RE ID 



balanced by the cohesion of the lower ice-layers, as there is no 

 friction against its bed. 



The empirical rule does not cover the case of a glacier like 

 the Mer de Glace, where the bed becomes steeper near its end, 

 causing a more rapid motion. But the law of flow must hold ; 

 here again the effective density of the ice is reduced by the open- 

 ing of wide crevasses, and the sectional area is undoubtedly less 

 than it would be if the slope of the glacier's bed continued uni- 

 form. 



In the case of glaciers with beds of uniform slope, the veloc- 

 ity and flow increase and decrease together, though not in the 

 same proportion.^ If we had a glacier of indefinite length and 

 of uniform section, it is evident that the direction of flow would 

 be parallel with the slope, and the velocity along any line paral- 

 lel with the glacier's axis would not vary as we move along the 

 direction of flow. To calculate the forces acting on a section of 

 such a glacier we have only to consider the weight of the sec- 

 tion, the viscosity of the ice and the friction against the bed. 

 The motion consists of two parts: (i) the sliding on the bed, 

 of which we have hardly any knowledge; (2) the shearing of 

 the section by virtue of which A B (Fig. i) would be deformed 

 into A' B' . Let us call the velocity of a point under such con- 

 ditions the normal velocity corresponding to that particular form 

 and size of the cross-section. 



A. glacier of uniform section could not exist if there was any 

 melting ; for we have seen that in this case the flow would 

 become smaller as we go down the glacier, and with uniform 

 section this would require a decreasing velocity, which, however, 

 is not admissible ; for, from the supposed uniformity of the 

 glacier, every section would be subject to the same forces and 

 would move with the same velocity. Therefore, the slope of the 



' The remainder of this section and all of the next are based on the theory that the 

 ice of glaciers acts in general like a viscous substance, and that the sole cause of 

 the motion is the weight of the ice itself. I am not discussing the physical nature of 

 ice which gives it this viscous property. Similar conclusions might be reached on 

 other theories. The rest of the paper is only dependent on the observed phenomena 

 of glaciers. 



