The Theory of Glacier motion. 137 



fluid particles over each other, thus adapting the ice to its 

 new shape, but the crushing by force of the ice into an 

 incoherent state, and then the process of refreezing it into a 

 solid mass. 



Tyndall himself describes it as a process of bruising and 

 regelation, and denies that ice is viscous at all. On the 

 contrary, he says emphatically, that a mass of ice at 32° is 

 very easily crushed, but it has as sharp and definite a 

 fracture as a mass of glass {id. 551). 



I am bound to say that I can find nothing in the 

 machinery by which glaciers are formed at all analogous to 

 these experiments, except the initiatory stage, when damp 

 snow is converted into continuous neve by pressure. From 

 that point down to the foot of the glacier I can find evidence 

 only of the ice being more and more condensed as the 

 pressure is exerted upon it by its walls, until it acquires the 

 •character of blue ice. I can nowhere trace evidence of the 

 pressure being sufficiently great, and such as to crush the 

 ice into powder, or to assimilate it in any way to the dis- 

 integrated condition which ice assumes in a Bramah press. 

 When the pressure is the greatest, and the blue veins are 

 produced, that is, when the great mass of the ice is detained 

 by some projecting obstacle, there we have the least sign of 

 •disintegrated ice. On the contrary, there the ice is most 

 transparent, and where the ice is embayed, and has there- 

 fore to spread itself and accommodate itself the most, then 

 the signs of internal strain in the form of blue bands are 

 •conspicuously absent. 



Similarly, if we examine the ice at the base of a glacier 

 by entering one of the well-known ice caves, instead ot 

 finding its internal structure filled with cloudy lines, and 

 more or less opaque, as Tyndall himself shewed to be the 

 case when incipient crushing is in progress {Glaciers of the 

 Alps^ 409 and 10), we find the ice to be most blue and 

 most transparent. On this matter Professor Bonney writes: 



