THE CONSTITUTION OF THE COPPER-TIN SERIES OF ALLOYS. 
15 
composition H and the temperature is 400°. At this temperature the r] becomes less 
stable than the body H, and the reaction 77 -f liq = y commences. 
The reaction is closely analogous to that characteristic of the C point in the 
copper-rich alloys, and like it is a very slow one. The H at this temperature has a 
composition very near Sn 45 ; the 77 is still almost identical with CugSn. This 
reaction should complete itself isothermally until, for alloys to the left of H' the 
liquid, and, for alloys to the right of H', the 77 is wholly transformed. But as soon 
as the plates of 77 have become completely coated with H, the reaction, which is now 
a process of diffusion through the solid H, becomes extremely slow, and, with 
ordinary rates of cooling, it never reaches anything like completion. The result of 
this imperfect reaction is that the diagram contains four compartments in the space 
EgSUH, in each of which one of the phases that are found would be absent if the 
equilibrium transformations had been completed. Professor Boozeboom has sug¬ 
gested the excellent plan of putting a bracket round the symbol of the phase that 
has no right to be present. We retain this notation because it indicates the ordinary 
condition of the alloys as most observers will find them ; we have, however, succeeded, 
by maintaining the ingots for many days at a temperature a little below 400°, in 
completing the reactions and removing the third phase. 
(9.) The HI alloys, containing from 87'5 to 98‘3 atomic per cents, of tin, that is, 
from 93 per cent, to 99H per cent. l)y weight. In these alloys the solid first formed 
is II, and the diagram sufficiently explains itself 
( 10 .) The IK alloys, containing more than 9 8 ’3 atomic per cents, of tin. These 
alloys contain primary crystals of pure, or very nearly pure, tin in a eutectic made 
up of H and tin. 
SECTION II. 
On the Solidification of a Metal. 
Our experience, and we believe it to be identical with that of most who have 
studied the subject, points to the comparative rarity of the formation of large 
crystals with plane faces during the solidification of a metal or alloy. The first solid 
structure is generally a crystal skeleton, which in its simplest form consists of a stem 
with radial branches jirojecting from it. This may he compared to a fir tree, and in 
many cases the branches are at right angles to the stem and to each other. Such a 
structure, when cut by a plane, gives rise to the fern-leaf or dendritic forms so often 
seen on the surface of cast metal, or in the etched and polished surface of a section of 
an ingot. 
Let us think of the case in which such a crystal skeleton has been produced in an 
otherwise liquid mass of metal. As the branches give off what we may call twigs, 
and these may develop other systems of twigs, and so on indefinitely, it follows that 
