14 
MESSES. C. T. HEYCOCK AND F. H. NEVILLE ON 
greater or less breadth is j^artly due to their varying inclination to the plane of the 
section. These plates, the first appearance of the 77 phase, must he either pure CugSn 
or solid solutions of Cu^Sn and CugSn ; we are not at present able to decide this 
point. Thus immediately Ijelow D'E' the alloys are a complex of 17 and residual y. 
But Roberts-Austen and Stansfield have proved that wlien any of the DE alloys 
fall to the temperature D' they evolve heat (see cooling curves of 20 , 21 , and 22 ). This 
must he due to the conversion of the residual y into 8 , so that below D'E" the alloys 
form the complex 8 + 77 (Plate 5, fig. 56). 
( 6 .) The EE alloys, containing from 25 to rather more than 27 atomic per cents, of 
tin, that is, from 38’4 per cent, to 41 per cent, by weight. These go through the 
same stages of y fi- liquid, then homogeneous y, then y + 77 ; but when the 
temperature of the line fG is reached, the residual y breaks up into 77 and liquid of 
the composition G. Thus these alloys present the somewhat rare phenomenon of 
the partial melting of a solid brought about by cooling it. The transformation 
Jf = Vty ■P 014 as we may safely write it, jf = CugSn + liq^, is the transformation 
which causes the angle at G in the liquidus. It is accompanied by a large evolution 
of heat, well seen in the cooling curve of Sn 27. 
It may be noted here that the triangular area l^f forms a region of uniform solid 
solutions, which could only have been discovered by the examination of chilled 
alloys, inasmuch as these solid solutions Ijreak up into two phases when slowly 
cooled. 
( 7 .) The EG alloys, containing from 27'5 to 42 atomic per cents, of tin, that is, from 
41 per cent, to 57’5 per cent, by weight. These, like the preceding, begin to solidify 
by forming the complex y + liq (figs. 75 and 76), their state, when the G temperature 
is reached, being that of y crystals of the f percentage and liquid of the G percentage. 
The isothermal transformation y = 77 + liq now commences, and completes itself 
abruptly with a heat evolution very well marked in the alloys near f. The alloys 
become perceptibly more liquid as the temperature falls,and the microscope shows 
very well the change from the rounded, uniform y combs to the comijlex of liquid 
between plates of 77 (figs. 73 and 74). 
When an EG alloy has cooled IdcIow the G temperature the 77 continues to 
crystallise out of a liquid which is continually becoming richer in tin. This process 
goes on between the G temperature of about 630'^ and the H temperature of 400°. 
Below 400° the EG alloy^s follow the same course as the next group. 
( 8 .) The GH alloys, containing from 42 to about 87'5 atomic per cents, of tin, that 
is, from 57'5 per cent, to 93 per cent, by weight. When these alloys begin to 
solidify they deposit plates of 77 , and this process continues until the liquid has the 
* This peculiarity was noticed during the operation of chilling the ingots : a little ingot of, say, Sn 28, 
chilled just above the G temperature, was not distorted by the sudden immersion in water employed to 
chill it, but a similar ingot chilled a little below G was converted into a mass resembling granulated zinc; 
this difference was observed more than once. 
