292 
PROFESSOR J. A. EWING AND MR. W. ROSENHAIN 
case, the brightness shades off across individual crystals, sometimes uniformly, so as to 
give the crystal a concave appearance ; at other times in patches, giving a mottled 
effect. On closer inspection, it appears that such variations in brightness occur only 
in one direction in each crystal; in that direction the brightness may be made to 
move across the crystal by a slight oscillation of the plate, but another direction can 
generally be found where the crystal appears uniformly bright and is quite free from 
the concave appearance. 
It is well known that the usual selective effect of oblique illumination on etched 
crystalline surfaces is due to the fact that etching develops on the surfaces of 
crystals a great number of minute hut similarly oriented facets or pits, and the 
uniformity of illumination over any area is a direct result of the constancy of 
orientation of these pits over that area. In the specimens of tin-plate here described 
we would therefore expect to find that the orientation of the etched pits or facets 
was not strictly constant over the entire area of each grain. 
By examining the etched surface of tin-plate under the higher powers of the 
microscope the etched pits or facets can he clearly resolved. Their appearance is 
illustrated at a magnification of 100 diameters in figs. 29 and 30. Under this power 
their orientation is apparently quite constant, and even under a power of 1000 
diameters the departure from strict constancy is only very slight. That it does exist 
could only he seen by comparing the outlines of apparently similar pits at opposite 
ends of a large crystal; even there the change in the angles of the figure or of the 
orientation of one of the sides, as compared by means of a cross wire in the eye¬ 
piece, was too small to be measured, but in certain cases the character of one of the 
outlines of the pit altered slightly, being a fine line at one end of the crystal, and 
a narrow wedge-shaped dark area at the other. This appearance indicates a change 
in the slope of that side of the pit, and thus a change of orientation of the internal 
facets of the pit. This change of orientation seems therefore to occur principally— 
if not entirely- in the vertical plane, i.e., in the plane of least thickness of the 
layer of tin. 
Under these circumstances it seems that the term “crystal” should only be used with 
some qualification in speaking of the patches developed by etching the surface of 
commercial tin-plate. In a crystal which is undistorted by elastic strain constancy 
of orientation of the elements is an essential characteristic of a true crystal. Some 
light is thrown on this matter by the fact that we have not observed this “concave” 
appearance in samples of solid tin, but only on commercial tin-plate where we are 
dealing with a very thin layer of tin adhering to sheet-iron. Even in tin-plate this 
effect may be made to disappear by re-melting the tin over a flame and allowing it 
to cool slowly and at rest. The appearance produced on re-etching such a specimen 
is shown in fig. 31 (to the same scale as fig. 28). Here, although the crystals are 
marked with a multitude of straight lines, suggesting a great number of twin 
crystals, there is little or no trace of the “ concave ” effect seen in fig. 28. 
In view of these observations we are inclined to attribute the “ concave ” appear- 
