THE CONSTITUTION OF THE COPPER-TIN SERIES OF ALLOYS. 
19 
transformation. In fact, when the cooling is spontaneous, that is, brought about by 
extinguishing the gas and allowing the ingot to cool in the furnace, although the 
total time spent in cooling may he considerable, yet the rate of cooling at the higher 
temj)eratures will always be rapid. We therefore made a series of chills, preceded by 
a cooling which was very .slow from the freezing-point of the alloy down to the 
moment of chilling. We took esj)ecial care to ensure that the time of cooling 
through a range of a few degrees above and below the critical point C or D should 
last several hours. In these chills, which we call “ slow-cooled chills,” and sometimes 
indicate by the letters “s.c.c.,” the detail formed before the chill is naturally large 
and the transformations at the critical points are much more complete.* This slow 
and regular cooling was obtained by carefully regulating the gas pressure and at the 
same time slowly and automatically cutting off the supply. (The arrangements for 
the automatic reduction of the gas supply are described in a paper in the ‘ J ournal of 
the Chemical Society,’ 1898, p. 714.) Some of these slow coolings required watching 
for twenty-four hours, and we are much indebted to Mr. W. Fearnsides for carrying 
them out. They would have been impossible without a temperature recorder such as 
that of Professor Callexdar, which we employed. 
On the Melting Point and the Rigidity of the Alloys at High Temperatures. 
Although the jjhotomicrographs of the successive cliills of an alloy enable us to 
determine with considerable accuracy the temperature at which the crystalline phase 
fills the whole volume of the ingot, this method gives us no information as to the 
solidity or rigidity of the solid phase at high temperatures. For example, in Sin, 
chilled at 1025° (fig. 2, Plate 1), la.rge primary combs of oriented and therefore 
crystalline sulistance certainly existed at the temperature of 1025°, and the chill at 
957° (fig. 3) shows that the crystals at this temperature occupy almost the entire 
volume ; hut the question whether at such high temperatures the crystals are rigid, 
like ordinary crystals, or plastic like the so-called liquid crystals, must be attacked by 
other methods, f Our attempts to determine the melting-point, as distinguished Irom 
the freezing-point, though unsuccessful in their primary object, which was to prove 
that the solidus was identical with tlie melting-iDoint curve, have estahlislied the 
great rigidity of the crystals, even at very high temperatures. 
AVe attempted to find the melting-points by two difterent methods. The first 
method was to subject a little bar of alloy to transverse strain while it was gradually 
* When the cooling, before the chill, was exceptionally .slow, the abbreviation v.s.c.c., for very slow 
cooled chill, is used. 
t The rounded form of the lobes of the primary skeletons in many of the upper chills suggests the 
action of surface tension and therefore of some plasticity in the crystals, but the continual transformation 
undergone by a crystal of solid solution as it grows may account for the rounding without the assumption 
that the crystals are plastic. 
