1890.] On the Plasticity of an Ice Crystal. 259 



VI. " On the Plasticity of an Ice Crystal. (Preliminary Note.)" 

 By JAMES C. McCoNNEL, M.A. Communicated by R. T. 

 GLAZEBROOK, F.R.S. Received May 29, 1890. 



Two years ago, in the ' Proceedings of the Royal Society,' was 

 published an account of some experiments on the plasticity of ice 

 made by Mr. Kidd and myself. We proved the oft-repeated state- 

 ment, that glacier ice is not plastic under tension, to be erroneous, 

 and, indeed, that an ordinary bar of ice composed of several crystals 

 will yield continuously either to pressure or tension. But we found 

 that a bar cut out of a single crystal with its length at right angles to 

 the optic axis showed no signs of continuous stretching, even when 

 subjected to half the breaking stress ; and other experiments con- 

 vinced us that an ice crystal will not change its shape under either 

 tension or pressure applied at right angles to its optic axis. These 

 results seemed to render it highly probable that an ice crystal was 

 not in any way plastic, and though, after the winter was over, we 

 wished that we had varied our experiments more, yet we quite 

 expected further experiments only to have corroborated the perfect 

 " brittleness " of a single crystal. 



Last winter I resumed the experiments alone. Cutting small bars 

 from uniform crystals, I supported their ends and hung weights half- 

 way between the supports. The result was the discovery of a peculiar 

 kind of plasticity in an ice crystal. The clearest idea of the nature 

 of this plasticity is given by the following analogy : A crystal 

 behaves as if it were built up of an infinite number of indefinitely 

 thin sheets of paper fastened together with some viscous substance 

 which allows them to slide over each other with considerable diffi- 

 culty ; the sheets are perfectly inextensible and perfectly flexible. 

 Initially, they are plane and perpendicular to the optic axis ; and 

 when by the sliding action they become bent, the optic axis at any 

 point is still normal to the sheet at that point. Thus, when a bar 

 with the optic axis transverse to its length is placed so that the axis 

 is horizontal, and the sheets of paper consequently vertical and longi- 

 tudinal, it refuses to take any plastic bend, however long the weight 

 be applied. If the bar be now turned over, so that the sheets of paper 

 are horizontal, quite a short interval suffices to produce a decided 

 permanent depression of the middle of the bar. In such a case, long 

 narrow bubbles in the ice originally vertical remain vertical, but the 

 optic axis bends with the bar, so that in one half of the bar it is 

 inclined to its position in the other half. The sides of the bubbles 

 were unbroken by steps or " faults," showing that the sliding did 

 not take place at a limited number of surfaces, but was an all- 



