CH. X.] ICE AND ITS WORK 157 



forming a straight line across the ice, as at a, they now form 

 a curve at b ; the stakes in the middle of the row have got 

 farther from a than those at the sides, and it is therefore 

 clear that they must have moved faster. But the movement 

 of the stakes is due simply to the movement of the ice, so 

 that if the middle stakes move faster than the side ones, it 

 shows that the middle of the glacier moves faster than its 

 sides. Exactly the same thing 

 may be observed in a river : light 

 bodies floated on a stream move 

 like the stakes carried down by 

 the glacier. Nor is it difficult to 

 see why a river should flow more 

 rapidly in the middle than at its 

 sides. The particles of water at 

 the sides rub against the banks, 

 and consequently are not so free 

 to move as the particles in the 

 middle of the stream. In like 

 manner, friction against the rocky •« - 



■ walls on the flanks of a glacier ^'^- 4°. -Motion of a glacier. 

 causes the ice at the sides to 



move more sluggishly than the ice in the middle. Again, 

 it is known that, in a river, the particles at the bottom drag 

 along the bed and move less rapidly than those at the 

 surface. The ice of a glacier behaves in precisely the same 

 way. It is concluded, therefore, that the motion of a glacier 

 is like the motion of a river. If the glacier enters a gorge, 

 it becomes contracted and the flow is rapid : while, if its 

 bed widens, it spreads out and the movement becomes 

 slower. In truth, in all points, the motion of a glacier 

 resembles that of a river; the movement is essentially 

 the same in kind but different in degree, the rate of 



