340 



Messrs. J. C. McConnel and D. A. Kidd. [June 21, 



and 



( 14- ®( v b—Va)M \ 



a 3 + o? 2 



(.Vb — Va) Ct. 



The expression — — - is a maximum when x = ft, and then we have 



by (i)- 



lr'-r _ Vb — Vg 



r~Tt 2a~ (2.) 



When St is very small, the ratio of r' — r to ht is the rate of increase 

 of the distance between A and B. So, if we take any two points of 

 the glacier at unit distance, the rate of increase of the distance 

 between them will be greatest when the line joining them is at 45° 

 to the direction of motion, and this maximum value will be equal to 

 one half the difference of the velocities of two points situated abreast 

 of each other and also at unit distance. 



Thus the maximum rate of extension in the case we have taken on 

 the Rhone glacier is 0*0029 mm. per hour per length of 10 cm. This, 

 be it remembered, is the most rapid extension selected from a large 

 number of measurements on different glaciers and at different times, 

 and yet only one of the three specimens of glacier ice showed a rate 

 less than this, and that was under one-third of the breaking tension. 

 The larger the specimen, the greater average plasticity would it dis- 

 play ; for the addition of a small piece like our second specimen, for 

 instance, would suffice to make a long rigid bar appear very plastic. 

 Hence the glacier itself would be far more plastic than most small 

 specimens taken at random from its mass. It would seem, therefore, 

 that neither the presence of crevasses nor a thawing temperature are 

 essential conditions of the motion of a glacier. But that crevasses 

 are found is not surprising, when we consider the rotten state of the 

 ice during the summer and the certainty that a crack, however small, 

 once formed will continue as long as the tension exists. We believe 

 further that the stresses produced in a glacier by its own weight are 

 comparable with those employed in our experiments. 



Description of Apparatus. 



We had two sets of apparatus in operation. The first was that 

 employed by Dr. Main, figured and fully described in his paper. 

 We reproduce his figure unaltered, though we made a few alterations 

 in the surrounding boxes. As we expected at first that our chief 

 difficulty would be keeping the temperature constant, we made special 

 arrangements for overcoming this. To secure a large heat capacity we 



