ox THE CRYSTALLINE STRUCTURE OF LEAD. 2.37 
quicker does re-crystallisation proceed. In some specimens which had been only 
slightly strained no visible change was seen until a fairly high temperature was used. 
One specimen was strained in tension and annealed for 20 hours at G0° Ct without 
any visible change taking place. When, liowever, it was annealed for two hours at 
100° C., the re-crystallisation seen in hg. 24 occurred. It is interesting to notice the 
large size of the patches showing new orientations in this photograph. 
An important question now presented itself for solution by experiment. Is the 
re-crystallisation which is apparent immediately after etching in a severely-strained 
crystal a direct and instantaneous effect of the strain, or is it a growth which occurs 
during the interval of time that has elapsed between the straining and the examina¬ 
tion ? The experiments of Ewing and IIoseniiain showed that a slow progress of 
growth goes on at atmospheric temperature in ordinary lead after sevei'e straining, 
which may result in the formation of comparatively large crystals in a severely 
crushed specimen after the lapse of several days or weeks, and the experiments just 
described show a similar slow change. In the present instance we are concerned 
with individual cry.stals in a structure of much coarser grain than was dealt with in 
their’ experiments, and with the comparatively short interval of time (some five 
minutes at least) which was required to prepare the specimen for examination by 
etching after the strain had been applied. At first it was not suspected that the 
re-arrangement of crystals seen after straining was other than an immediate effect 
of the strain, but the author has now satisfied himself that this is not tlie case. The 
re-arrangement does not occur in the act of straining like the re-arrangement (by 
twinning) which one can produce on straining a crystal of calcite. It occurs after the 
strain has taken place, during the time that elapses before the crystal is etched for 
re-examination, and though it requires only a short interval of time for its develoji- 
ment, it is to be classed with the progressive growth demonstrated by Ewing and 
RosENHArN and confirmed by the experiments already described in this paper. 
That this was the case was first suspected from observations of the character of 
the re-arranged crystals. If these had been produced by successive twinning actions 
forming a direct result of the strain, as in the twinning of calcite, we should have 
expected to find tlie straight-line boundaries characteristic of twins, not only between 
the patches having new orientation, but also between these and the unchanged 
portion of crystal in which its original orientation was preserved. Now, although 
there are numerous straight-line boundaries between the patches of new orientation, 
demonstrating the twin relation of these patches to one another, it is remarkalfie 
that twin boundaries are not to be found between the unchanged portion of the 
crystal and any of tlie re-crystallised portions. Twins to the origlual orientation of 
the crystal would undouljtedly he formed if the action was the direct result of 
straining, such as lias been described above, but in all cases the boundaries between 
any new patches and the original structure were found to be irregular in outline, and 
to be eaten out into channels by the etching solution ; and no traceable geometrical 
