and Soft States in Metals. 269 
30. The change of colour and reflecting-power in gold is 
quite as remarkable as in silver ; and my observations on the 
changes produced on the surface-layers of these and other 
metals in the more massive forms serve to connect the phe- 
nomena in the attenuated and the massive metal with the 
same causes. 
31. In gold-leaf the state of strain which is left after flowing 
is always associated with the green translucence and the yellow 
metallic reflexion of burnished gold. 
Llectrical Properties. 
32. The electrical conductivity of a silver wire in the 
annealed state is 8 to 10 per cent. higher than that of the 
same wire in the hardened state. With other metals also the 
annealed state has the higher conductivity *. 
33. Mirror-silver, in which the metal is in the hardened 
and strained state, shows an initial potential difference of 
0*1 volt against ordinary silver in dilute acids+. Liidtke, 
who investigated this subject, regarded mirror-silver as an 
allotropic and colloidal modification of silver. 
34. My own observations{ have shown that a thermo- 
junction consisting of a hardened and an annealed wire of 
silver has an E.M.F. of 0:17 micro-volt for 1° C. of difference 
between the hot and cold junctions. This is a much higher 
H.M.F. than I have found recorded by other observers (see 
par. 64). At 260° the H.M.F. falls to zero, as the hardened 
wire then passes into the soft condition. Copper, iron, pla- 
tinum, magnesium, all show a measurable E.M.F. between 
the two states (see par. 64). In each of the above cases 
there is a well-marked difference between the two states, but 
the loss of conductivity through hardening seems to me 
peculiarly significant. 
Thermochemical Properties. 
.389. In etching hardened or polished silver it is found that 
solvents act first on the hardened phase, removing it and 
leaving the crystalline phase exposed. 
If the hardened metal is merely a slightly compacted form 
of the annealed, it would naturally be expected that the more 
open form would be the more readily attacked and dissolved. 
On thermochemical grounds Berthelot § regarded hammered, 
crystalline and annealed silver, as wellas certain other forms, 
* ‘ Practical Electricity,’ Ayrton, p. 157. 
+ Ludtke, Ann. Phys. Chem. (2) vol. 1. pp. 678-695. 
| * Nature,” vol. Ixx. p. 31. 
§. Berthelot, Compt. Rend. vol, cxxxii. pp. 234-241 (1901). 
