142 Canon Moseley on the Mechanical Impossibility of 



ration, and therefore the time of shearing, would be dependent 

 on the form and dimensions of the section sheared and the unit 

 of shearing. The unit of shear in this case could only be deter- 

 mined by observing the time; and I have given a formula for 

 so determining it in the case of a prism having a rectangular 

 section*. In a temperature above freezing, the prisms of ice I ex- 

 perimented upon could not but have been in the act of melting, 

 and the whole section of shearing in the act of diminishing by 

 thus being melted ; time must therefore in this way have had 

 an influence on my determination of the unit of shear. Heat, 

 too, must have been continually conducted to the ice through 

 the apparatus; and the process of shearing must have been ac- 

 celerated by the excess of pressure sustained at the upper and 

 lower surfaces of the prism, and the lowering of the temperature 

 of congelation, and therefore the melting at those points conse- 

 quent thereon. The section of the ice-cylinder was therefore 

 probably less than I have assumed it to be, and the unit of shear 

 greater. 



Mr. Ball says very truly that " in some substances the amount 

 of resistance opposed to the separation of adjoining particles is 

 nearly independent of temperature, and of the time during which 

 the pressure is applied. In other bodies, which oppose a very 

 considerable resistance to fracture, the particles gradually change 

 their relative positions under the prolonged action of even a 

 slight pressure ; and in some of them the change depends very 

 much on their temperature while exposed to pressure." Apply- 

 ing these remarks to my own determination of the unit of shear 

 in ice, I beg to remind Mr. Ball that it was made on one of the 

 hottest days (August 24) of the year 1869, when the thermo- 

 meter stood in the shade at 74°-75°, and that the ice I used was 

 almost in a state of deliquescence. No one who has inquired 

 into the temperature of glaciers has ever thought, I believe, of 

 assigning to it a temperature greater than this ice must have 

 had. 



Ice covered compactly with other ice, as it is in ice-houses, 

 does not melt as this did. If the covering be perfect, as it is in 

 a glacier, it probably does not melt at all. So far as the influ- 

 ence of temperature is concerned, the unit of shear in glacier- 

 ice, however little below the surface, cannot, therefore, but be at 

 least equal to that of the ice in my experiments; far down in 

 the interior of the glacier it is probably much greater. I have, 

 however, brought this question of the shearing of ice in a tempe- 

 rature below freezing, and of the influence of time upon it, to 

 the test of a direct experiment. On the 15th of February, 1870, 

 I placed a cylinder of ice, one inch and a half in diameter, in a 



* Phil. Mag. January 1870. 



