1912-13.] Application of the Law of Torsional Oscillation. 189 
Repeated experiments gave the values : 
and 
A. a = 0 
n = 2’30 
A. a= 1 
n= 1-70 
B. a = 50 
n = -40 
B. a = 20 
w = *40 
C. a = 50 
n = *60, 
C. a = 25 
n = ’50. 
Although no exact record of the temperature obtained was got, the 
wire in the latter case was raised to a much brighter heat than in the 
former two cases. 
German silver wire raised to a red heat and quenched gave values of 
n far higher than any got formerly for the same metal ; whilst two distinct 
lines were required, a being equal to zero in both cases, and n having the 
values 2*40 and 2 25 for large and small oscillations respectively. 
A specimen of iron wire, tested in the original state, gave the results : 
A. a = 0 and B. a =15 
n = 2*25 =*60. 
On raising to a high temperature (bright redness) and subsequently 
annealing, both values of n were found to be lowered, whilst another length 
of the same specimen, when raised to approximately the same temperature 
and quenched, gave similar values. The values of n in the first case were 
1*60 and *33, and in the second case 1*80 and *25. 
Work on Alloys. 
With rise of temperature, brass has been shown (see former paper) to 
behave in a way which contrasts strongly with aluminium. In the case of 
brass, one value of n, and thus one straight line, is got at the ordinary 
temperature ; this one line, although changing in inclination to the axis, 
persisting up to a limiting temperature in the region of 370° C. When this 
temperature is reached the one line breaks into two, and two values of n 
are always got thereafter on further increase of temperature. With 
aluminium, on the other hand, two values of n are observed at the ordinary 
temperature, these two values gradually converging with rise of tempera- 
ture, and finally coinciding at the same limiting temperature as before. It 
was thought, then, that an alloy of the two metals might give interesting 
results, and that temperature changes might be got which would prove 
interesting in relation to those already obtained for the separate 
constituents. 
Three separate brass-aluminium alloys were made ; the brass was 
