286 
PROFESSOR J. A. EWING AND MR. W. ROSENHAIN 
The low power required is a further advantage in view of the fact that an etched 
lead surface must be kept wet with an acid solution while it is being examined and 
photographed, as such a surface becomes dull and useless as soon as it dries. 
To identify the areas which were to be kept under observation, we marked them 
by scratches or furrows made witli a steel point. Unsymmetrical patterns of marking 
were used for this purpose to facilitate replacing the specimen in its former position 
and orientation. It is an essential precaution in taking micro-photographs under 
oblique illumination which are to serve for purposes of comparison, that the 
orientation of the specimen and the angle of incidence of the light should be 
constant throughout the entire series. Our micro-photographic arrangements 
enabled us to comply with these conditions to a considerable degree of accuracy. 
The first series of these midro-photographs (figs. 8 to 13 inclusive, Plates 5 and 6) 
was taken from a specimen of lead which, after being severely strained, was simply 
kept during six months at the temperature of an ordinary room without any special 
thermal treatment whatever, being carefully re-etched and photographed at intervals. 
The first of the series (fig. 8) was taken immediately after the specimen had been 
strained. The magnification (12 diameters) is too low to clearly resolve the minute 
crystalline structure which existed at this stage. The identification marks are very 
clear at this stage, but they become less clear after each successive etching. 
Fig. 9 represents the same surface, re-etched after six days; a small amount of 
change is visible, and this becomes more marked in figs. 10, 11, 12, and 13, which 
represent the state of the specimen after one, two, four, and six months respectively. 
Comparing figs. 8 and 13, the great change in the crystalline structure is strikingly 
evident. 
Measurement of the largest crystals seen in fig. 13 gives some rough idea of the rate 
at which these crystals have grown. In one case the dimensions are as follows :— 
Length of crystal in freshly-strained specimen = •0025" 
,, ,, specimen after 1 month = '0083" 
,, ,, ,, ,, 4 months = '0100'' 
„ „ „ „ 6 „ = -0115" 
From this it would appear that the rate of growth is greatest in the first month 
after the specimen has been strained ; but a close examination of such specimens 
reveals a feature which considerably affects this question. It is seen that the crystals 
do not grow by the steady accretion of layers all over their surface, but that they 
throw out arms or branches, which invade neighbouring crystals, thus forming; a 
skeleton crystal somewhat similar to those that are often noticed in the crystallisation 
of fused substances. The more or less dim and mottled appearance of the larger 
crystals in figs. 8 and 9, and the much greater brightness and uniformity shown by 
the same crystals in figs. 10, 11, and 12, is undoubtedly due to the formation of these 
skeletons, and their subsequent filling in. 
