and Soft States in Metals. 273 
amorphous state, it must be remembered that all the obser- 
vations on these properties have necessarily been made not on 
the pure A phase, but on mawtures of the two phases. In 
these observations the structwre which has resulted from the 
particular mode of hardening has probably had an important 
influence on the results obtained. This remark applies also 
to observations on the electrical and to the other mechanical 
properties. In every case, however, it can be said that as the 
transformation into the A phase has been pushed further and 
further, the special properties associated with it have become 
more and more marked. | 
64. The transition temperatures of silver and other metals 
were determined by a softening test. The metal was beaten 
into a thin elastic strip, and was wrapped round the bulb of a 
thermometer which was plunged into an air-bath of which the 
temperature was gradually raised. From time to time the 
strip was removed and examined for signs of softening. The 
lower point of the transition range was taken as soon as any 
sign of softening was observed. Heating was then continued 
till the strip became so soft that it could be laid perfectly flat 
on a glass plate by a slight pressure with the finger. The 
following results were obtained :— 
Silver began to soften at 230°, and was completely soft at 265°. 
Gold 3 Ss 250° ‘ ‘ 280°. 
Copper _,, a 250° " e 20g: 
Magnesium re 300° ‘ a 360°. 
The transition temperature of silver by thermo-junction 
method was 260°. By quickly raismg the temperature of a 
thermo-junction of hardened and annealed silver to a high 
temperature, a much higher H.M.F. was obtained than that 
givenat the transition temperature. Calculating this E.M.F. 
into degrees, the temperature to which the junction had been 
raised was 750°. This simply shows that, as is usual in the 
transformations of solids, the element of time plays an 
important part. 
Thermo-junctions of some other metals heated in this way 
gave the following maximum E.M.F.:— 
Copper, 32 ; iron, 800; and platinum, 24 micro-volts: 
In 1899 Dr. Magnus Maclean made a series of observations. 
on the effects of strain on the thermo-electric qualities of 
metals *. With copper soft and hard the H.M.F. varied from 
00027 to 0:0264 micro-volts per degree. Steel gave 0°1028; 
nickel, 0°3784 ; lead, 0°0076 micro-volt. These observations 
* Proc. Roy. Soc. vol. Ixiv. p. 322, and vol. Ixvi. p. 165. 
Phil. Mag. 8. 6. Vol. 8. No. 44. Aug. 1904. ge 
