the Descent of Glaciers by their Weight only. 141 



we may conclude that the unit of shear in glacier-ice is not less 

 than that given by my experiments, but greater than it. 



Mr. Ball objects to my assuming the resistance of glacier-ice 

 to be the same as that of the ice used in my experiments, which 

 was the ordinary ice of commerce collected on the surfaces of 

 rivers and lakes in Norway. " Glacier-ice/'' he says, " however 

 slight the indications of it on the surface, is a congeries of sepa- 

 rate fragments more or less perfectly welded together, and show- 

 ing by the frequent presence of air-bubbles, and by its behaviour 

 when exposed to radiant heat, an inferior degree of solidity w *. 



But this is precisely the kind of ice with which some of my 

 experiments on shearing were made. The ice was broken in 

 pieces and then hammered into the cylindrical hole of the shear- 

 ing- apparatus f, so as to be formed by regulation into a solid 

 cylinder. It was then of the consistency of the glacier-ice de- 

 scribed by Mr. Ball. It was visibly " a congeries of separate 

 fragments more or less perfectly welded together." The glacier- 

 grain of which M. Heim speaks was clearly seen in it. Its re- 

 sistance to shearing was nevertheless almost identically the same 

 as that of a cylinder of ice of the same diameter turned in a lathe 

 out of a block of Norway river- or lake-ice J. The adherence of 

 the new surfaces brought into contact in the act of shearing was 

 as perfect in this granular ice as in that cut from the block ; 

 and there was the same perfect continuity of the substance of 

 the ice§. 



As regards the influence of time on my determination of the 

 unit of shear, I must have expressed myself imperfectly, for I 

 have been misunderstood. In saying that it " cannot be deter- 

 mined with absolute accuracy without taking into account the 

 time of shearing" ||, I meant that in the act of shearing a prism 

 of ice in my apparatus, the dimensions of the surfaces which re- 

 mained unsheared were continually diminishing, and with them 

 the whole resistance to shearing ; so that to keep the conditions 

 of the experiment always the same — the shearing-force being 

 always just in excess of the resistance — the load must be con- 

 tinually diminished. The velocity of shearing would then be 

 constant. Or if, as in my experiments, the load remained the 

 same, whilst by reason of the constant diminution of the surface 

 of adherence the resistance diminished, the velocity of shearing 

 would be continually accelerated ; and the degree of this accele- 



* Phil. Mag. S. 4. vol. xli. p. 85. 



t Phil. Mag. January 1870. | Ibid. 



§ I believe this adherence of the new surfaces brought into contact in the 

 act of shearing, which is so conspicuous a quality of ice, has been observed 

 also in lead. 



|| Phil. Mag. January 1870. 



