591.] On the Plasticity of an Ice Crystal. 3'55 



Into this matter I enter more fully below. 



The indicated rise between 12.59 and 2.3 P.M. is, I feel sure, simply 

 lue to a misreading. Whenever the weight was altered the appa- 

 itus was unavoidably disturbed, so I had to take an entirely fresh 

 reading of the pointer. Generally this only differed by fractions of a 

 millimetre from the previous reading, but in the case in point it was 

 nearly 6 mm. greater. The ice showed an inconvenient tendency to 

 slip backwards on the iron plate, thus bringing the end of the 

 pointer forwards till it almost touched the edge of the slit. The ice 

 had to be pushed forwards three or four times during the experiment. 

 Of course a fresh reading was taken after each such displacement, so 

 that no error resulted. This trouble was caused doubtless by the 

 plate not being accurately level. In subsequent experiments I was 

 more successful in avoiding it. 



Exp. 11. I desired to establish with, the more delicate system 



measurement that the plasticity is inappreciable when the bending 

 3ss is applied at right angles to the axis. I cut a bar, all one 

 crystal, from the bath ice, and planed it so that the upper and lower 

 surfaces were as accurately as possible parallel to the optic axis. In 

 the polariscope, when the middle of the black cross was in the middle 

 of the bar, the two faces were equally inclined to the lines of sight. 

 I then set up the apparatus in the usual way. The results are seen in 

 the annexed table (p. 334). 



f It will be seen that the pointer indicated a rise of the stirrup 

 amounting in the 21^ hours to 0'29 mm. As was before mentioned, 

 the stirrup was slightly roughed to prevent it from slipping, so at 

 first it would make contact with the bar at only a few points. 

 Evaporation would help to extend the contact to large surfaces, and 

 admit of a slight movement of the stirrup relatively to the ice. 

 Thus the experiment was not as satisfactory as could be wished. It 

 is possible that a very slight depression of the bar might be masked 

 by this effect of evaporation. But even supposing that the rate of 

 real depression was twice as great as that of the apparent elevation, 

 viz., O0043 mm. per hour, it would still be very small compared with 

 the rates of the next experiment. I am at any rate entitled to say 

 that within the limits of error of experiment there is only one kind of 

 plasticity in an ice crystal, viz., that due to the sliding layers at right 

 angles to the optic axis. It is probable that the same source of error 

 was active in other experiments, but in them the effect would be 

 almost negligible. 



Exp. 12. The same bar was turned on its side so that the optic 



cis was vertical. 



