﻿M. A. Heim on Glaciers. 503 



constructed : their situation on the Rhone Glacier is indicated 

 by thick lines in fig. 10 : — one in the region of A ; the other on 

 the other side, above. 



In a somewhat more complicated experiment with killed gyp- 

 sum, in which two larger and one smaller streams were brought 

 to confluence, longitudinal lines also were interrupted by dis- 

 placements. The middle wall of the two larger streams was 

 cut through obliquely by fine fissures at which the gypsum was 

 slightly displaced vertically. In one set of experiments the 

 cracks corresponding to the usual glacier-fissures, which go ob- 

 liquely upward from the margin toward the centre, were very 

 distinct ; but instead of being very open, they were more nume- 

 rous. In other experiments they appeared to be rendered un- 

 necessary by displacements in the mass. An attempt to photo- 

 graph all the fine forms must be repeated. 



Expansions. — When a glacier, in consequence of entering a 

 wider part of its valley, can expand at its lower end, at the upper 

 part of the widening it throws out longitudinal fissures which, 

 curving to the borders, form at the lower end a radiating, fan- 

 like system. This is readily intelligible. The pressure of the 

 middle portion, to which the ice can now yield, is radial, and 

 carries the ice in a periphery having the end of the glacier for 

 its centre ; the fissures are thrown out perpendicular to the di- 

 rection of motion, and are therefore radiating. Remarkably it 

 is not so in experiments with killed gypsum. There two sys- 

 tems of fine fissures occur at the end, which cut one another at 

 a constant angle of from 69° to 72°. These fine fissures never 

 gape. At them the lower parts are displaced vertically upward. 

 They are nearly similar triangles — one side formed by the border 

 of the out-spread mass, the two others by a fissure of each of 

 the two systems. The system of radiating gaping fissures in 

 the glacier is here replaced by two systems of displacement- 

 fissures which cross one another. 



In wet masses so small, adhesion is proportionally strong, 

 and more opposed to divisions than to displacements. We have 

 already above encountered this preference in the killed gypsum 

 for displacements, which appears to me to be the cause of the 

 difference here — though perhaps it has a deeper foundation in 

 the essential nature of the substance. Unfortunately I was 

 unable to investigate other semiliquid non-viscous substances in 

 this relation ; nor do I know how far the angle of about 70°, 

 peculiar to killed gypsum, depended on the degree of thick 

 liquidity. In glaciers I have not found any trace of a double 

 system, much as I have sought for it. Fig. 11 shows the course 

 of the fissures at the end of the Rhone Glacier ; fig. 12, the dis- 

 placement-fissures at the expanded end of a gypsum stream. 



2L2 



