54 
MESSES. C. T. HEYCOCK AND E. H. NEVILLE ON 
the chill did not prevent the formation of a few long slender bars of 77 , dark in the 
photograph. 
Sn 22 . S.c. chill at 680°, and chills at 650° (not reproduced). 
This is a remarkable example of the polygonal pattern produced by cutting the 
grains of solid solution composing the ingots at difierent angles, and so causing them 
to etch differently, though they are probably the same chemically. Tilting, or 
rotation under oblique illumination, does not much alter the relative shade of these 
patches, and they show no iridescent effect. The bars of 77 are much scantier than in 
the preceding ingot, hut they can he found. The chill at 650° is very similar to the 
preceding, and as far as we can see it is free from bars. Another chill at 650°, very 
slowly cooled before the chill, lias many more bars, but they are small and thin, and 
may, we think, he attributed to the chill not being sharp enough to arrest all further 
change. In fact, the bar pattern is so slight in all the above-mentioned ingots, that 
we think it almost certain that the alloy passes through a region of temperature in 
which it is really a uniform solid, as in the case of Sn 21 , 
Sn 22 . S.c. chill at 600° (not reiiroduced). 
We should have expected the bar pattern to become a real feature when the 
transformation cuiwe was passed, as in the present chill, hut it is, on the contraiy, a 
very good example of a uniform solid solution consisting only of a few large patches, 
which turn from dark to light when tilted. There is hardly a trace of the bars. 
Even the chill at 588° is very similar, though here some of the polygons are hounded 
by straight lines and may he massive bars. 
Hence, although the cooling curve of the alloy shows a considerable evolution of 
heat at 627° where the transformation curve is reached, yet the chills fail to show the 
corresponding change. 
Sn - 22 . Chill at 570° (not reproduced). 
The polygons, still very vaiied in tint, are now crossed and sometimes margined by 
massive bars of 77, but there is no lace either in the bars or the ground. 
Sn 22 . Chilled at 565°, a sloiu-cooled chill (fig. 60). 
This is certainly below the last, or I)', halt on the cooling curve, a point marked in 
the cooling curve of this alloy by a somewhat sloping flat. A lens now shows massive 
bars and curved bands of 77 dark on a light ground. A slightly higher power brings 
out the fact that this ground is full of paler bars and of lace bars, the effect being even 
better seen in fig. 61 . Thus the crystallisation has taken place in at least two stages: 
at first the massive 77 bars formed, and, secondly, the mother-substance broke up into 
