324 Mr. J. C. McCoiinel. [Mar. 12, 



tion would only have corroborated the perfect " brittleness " of 

 single crystal. 



Since our paper was written, my attention has been called to 

 passage in Professor James Thomson's masterly article on " Tbi 

 Lowering of the Melting Point of Ice by Distorting Stress 

 (' Phil. Trans.,' 1849), in which he expresses the opinion that crystals 

 whether of ice or other substances, are not plastic. 



If we reject the idea of internal distortion of the crystals, we an 

 driven to the conclusion that the observed plasticity must be due t< 

 some action at the interfaces, whereby the crystals alter their shapi 

 sufficiently to allow them to alter their relative positions. As to th< 

 nature of the action, various suggestions occurred to me. Jama 

 Thomson explained the plasticity of ice at C. by supposing the 

 to melt at those interfaces where the stress was great, and the libe 

 water, after flowing to points where the stress was small, to 

 solidify. This might be extended to low temperatures by supposi 

 a certain quantity of water to be kept in the liquid state by 

 pressure of residual impurities. But the process would be en 

 mously retarded by the constant necessity for the distribution of 

 being equalised by diffusion. Again, it is not clear how a bar of i 

 during this process would be able to resist a tension considera 

 greater than the pressure of the atmosphere. With more probabili 

 we may suppose one crystal to grow at the expense of another owinj 

 to the stresses and strains on the contiguous parts being differe: 

 Though the stresses were the same, the strains might be dift'eren 

 owing to seolotropic elasticity. But the elasticities are not likely 

 be very different in different directions, so for a very small exte: 

 of the bar we should expect considerable movement of the interfi 

 There is, however, nothing to prevent the stresses being different 

 The tension in any direction parallel to the interface might be greater 

 in one crystal than in the other. The migration of matter from one 

 crystal to another under less stress would probably in almost all cases 

 be accompanied by yielding to the external force producing the 

 stresses. But in this case the effect would be very indirect, and 

 again we might look for large movement of the interfaces. 



Some such speculations had occupied my mind last autumn, and it 

 was with considerable curiosity that I began experiments in Decem- 

 ber on the puzzling question of the real cause of the plasticity of ice. 

 I took a bar of ice consisting of half a dozen crystals, made a draw- 

 ing under the polariscope of the relative position of the interfaces, 

 and then set up the bar with the ends supported and a weight hung 

 from the middle. After two days, it had bent a good deal, yet, under 

 the polariscope, I could detect no material change in the position of 

 the interfaces. One crystal, however, had completely chang< 

 appearance. It now strongly reminded me of a piece of unaunealed 



