20 Properties of Plastic Crystals of Ammonium Nitrate. 



the crystals are brittle. It was not found possible to observe 

 the effect of a direct tension on the long crystals. 



Prolonged grinding, or heating in air to 30°-40° C 

 produces a white opaqueness. This probably indicates the 

 transition to Groth's second form, said to occur at 32° C 



On applying a load to a crystal supported horizontally, the 

 initial rate of bending is the greatest (except when fracture 

 is occurring). The rate of bending increases much more 

 rapidly than in direct proportion to the stress (for the loads 

 that were employed). This result, deduced from bending 

 experiments, is verified by the pointed shape of the curve 

 assumed by the crystal when bent. 



A pseudo-viscosity f tt ) (calculated as if the bending were 



viscous) gave the results of the order of that of glacier ice. 



Curves showing how the "initial" bend, (obtained immedi- 

 diately on applying a load) varied with this load were ol tained. 

 This initial bend may possibly be due to the narrow crystals 

 which exhibited it being complex in structure. 



Attempts to determine whether slip occurred parallel to 

 the length of a bent crystal indicated that if there were any, 

 it was not large. Simple slip in other directions could not 

 account for all the bent shapes obtained. The form consisted 

 in some cases of a sharp " "V " shape, the bending being 

 confined to a very small central portion of the long crystal. 



In bent crystals, ground and mounted, the extinction-lines 

 seen on viewing the specimen between crossed Nicol prisms, 

 are not normal to the curved crystal. These lines appear 

 always to be inclined at roughly 45° in either direction to 

 the radius of curvature of the crystal. The sense probably 

 depends on irregularities and the exact mode of bending. It 

 may change rapidly. There is a slight progressive change in 

 the direction of the plane containing the optic axes, on crossing 

 the crystal radially. To a first approximation, however, the 

 plane remains parallel to the length of the crystal. The 

 angle between the optic axes is little, if at all, changed by 

 the bending. 



On the whole, the structure, as indicated by. the optical 

 properties, is little changed by the bending. This may be 

 compared with the experiments of T. Terada and P. Carmak. 



It seems probable that (whether or not there be any 

 simple slip) a local rotational atomic adjustment occurs, 

 causing the crystal to remain in equilibrium, though bent. 

 This rotation would vary steadily in amount as the crystal is 

 crossed radially ; and it would probably be accompanied by 

 a rotation of the plane containing the optic axes, which would' 



