156 Messrs. ft. M. Deeley and P. H. Parr on 



the velocity would appear, as we hope to show, to be approxi- 

 mately proportional to the downward force of gravity ; but in 

 the case of a grinding tool, such as that of the lead-emery 

 surface, when slip occurs the motion is rapid and the coefficient 

 of friction is regarded as being due to the interlocking of the 

 asperities of the two surfaces, and the movement results in 

 abrasion. Also, in the case of polishing or grinding, unless 

 the stress be very considerable, the polishing medium does 

 not slip at all. 



We shall give grounds for holding that the motion or slip 

 of clean ice upon a hard, but slightly rough surface, may take 

 place without abrasion of either the rock or ice. According 

 to this theory what is known as slip differs in the case of a 

 glacier from our conception of the ordinary meaning of the 

 word. 



It is well known that the freezing-point of ice varies with 

 the pressure. An old experiment which has lately been 

 repeated by Tarr and Rich * illustrates this property of ice. 

 A loaded wire rests upon a block of ice. Owing to the 

 pressure exerted upon the ice immediately below the wire, the 

 ice is there melted, the water moves up the sides of the wire^ 

 and the pressure being relieved, freezes again above. There 

 is thus set up a continuous flow of water from the underside 

 of the wire to the upper side, and the wire cuts its way 

 through the ice without destroying the continuity of the 

 mass. In one of Tarr and Rich's experiments a block of ice 

 2*57 cm. thick was cut through in four minutes, the load 

 being 70*32 kg. per cm. 2 , and the temperature — o, 055 C. 



When metallic wire is used the heat is rapidly conducted 

 from the upper to the lower surface of the wire, and the 

 sinking of the wire through the ice is rapid. We shall show, 

 however, that when a slowly conducting material, such as 

 glass, is used the action takes place much more slowly. 



A smooth rock surface upon which a glacier rests may be 

 regarded as being covered by numerous small projections 

 which rise into the ice above. Against the up-stream sides 

 of these projections the pressure of the ice is greatest, whilst 

 on the down-stream sides it is least. On the up-stream sides 

 melting, therefore, takes place, the water produced passing 

 round the projections and freezing again on their lower sides. 

 In this way the mass of ice is allowed to move slowly forwards 

 over the somewhat rough surface. 



If we consider the projections on the glacier bed to be 

 pyramidal, with base angle 0, then the area of the base is 

 (2s) 2 , the area of the vertical cross-section through the apex 

 * Zeit.fiir GletscherJcunde, Band vi. Heft 4, p. 229. 



