266 Mr. G. T. Beilby on the Hard 
as allotropic modifications. He found that the heat of solution 
in mercury of hammered silver was + 2°03 cal., as compared 
with +0°47 cal. for silver heated to 550°, and +0°10 cal. for 
crystals of electrolytic silver. My observations show that 
550° is an unnecessarily high temperature for annealing silver. 
As shown by the thermo-junction (par. 34), the transition 
begins about 250°, and would probably complete itself at 
that temperature if sufficient time were allowed. 
36. This completes the summary of the evidence in sup- 
port of the view that the hardened and the annealed ave distinct. 
forms of metal. By collecting the hitherto scattered obser- 
vations on the mechanical, optical, electrical, and thermo- 
chemical properties of the hard and soft forms, it is seen that 
these all, without exception, support the view which [I first 
advanced nearly four years ago, that the micro-structure 
shows that there is a well-marked difference between the two 
states, the crystalline, and the amorphous or colloidal. 
37. The evidence from these various sources also shows 
how naturally the varied properties group themselves on 
either side of a transition-temperature which is common to 
all; we are therefore justified in regarding these forms of a 
metal as two distinct phases. 
A Phase Theory of Hardening and Softening. 
38. With the recognition of these distinct phases the 
softening of metals now for the first time takes its place 
among other phase-transformations; there are, however, 
certain special features which are peculiar to this trans- 
formation. So far as is yet known, the transformation 
from hard to soft is thermally irreversible; that is to say, 
no mere reduction of temperature below the transition-point 
is sufficient to bring about the reverse change from soft 
to hard. In the well-known case of the two forms of tin, the 
white and the grey, it may be found that the latter is amor- 
phous, not crystalline; but at present there is no evidence 
on this point ; and the fact that the white variety expands 
considerably in passing to the grey seems to place the 
transformation in a different class from that now being 
discussed. 
39. Though the transformation from soft to hard cannot be 
brought about by change of temperature, itis readily effected 
mechanically. Whenever work is done on the C phase, 
hardening takes place. The micro-structure shows that this 
hardening is always accompanied by the disappearance of 
crystalline and the appearance of vitreous and granular 
characters. Polished or flowed surfaces are covered with a 
a i i 
