90 Proceedings of the Royal Society of Edinburgh. [Sess. 
Improvement of the Magnetic Quality . — Before proceeding with any 
thermal treatment which might have a permanent effect on the materials, 
it was thought desirable to make some tests which could be used later to 
ascertain whether the heat treatment had probably resulted in change of 
constitution. The most suitable test for this purpose was to reinvestigate 
the magnetism of the alloys when they were cooled to the temperature of 
liquid air. Such a procedure seldom has any influence in altering the con- 
stitution of alloys. In cases where transformations do occur between +15 a 
and — 190° C., their velocity is so slow that the magnetic test can be carried 
out and the specimen brought back to room temperature before the con- 
stitution has been sensibly affected. 
Liquid Air Test on the Cast Specimens . — When the alloys were cooled 
to liquid air temperature and tested thereat in the usual manner,* the sus- 
ceptibility showed in most cases a more or less marked increase. The 
retentivity and coercive force were altered in a like manner. Table III. 
gives the percentage changes produced in the values of I and C.F. by the 
cooling. 
Table III. — Liquid Air Effects. 
Percentage Change. 
Test. 
Alloy. 
14 per cent. 
16 per cent. 
18 per cent. 
30 per cent. 
38 per cent. 
48 per cent. 
tin. 
tin. 
tin. 
tin. 
tin. 
tin. 
ho 
+ 15 
+ 5 
+ 15 ! 
0 
ho 
+ 17 
+ 12 
+ 20 
+ 2 
hoo 
+ 21 
+ 26 
+ 35 
+ 1 
T 
- l 200 
+ 29 
+ 34 
+ 60 
0 
T 
J -300 
+ 33 
+ 37 
+ 65 
+ 
0 
6 
T 
J-r 
+ 45 
+ 60 
+ 35 
... 
0 
C.F. 
+ 20 
+ 30 
+ 15 
+ 10 
+ 
The 38 per cent, and 48 per cent, alloys showed almost no change in 
susceptibility due to the change in temperature, but the coercive force and 
consequently the hysteresis loss were increased. In employing the values 
given in this table one must bear in mind that the changes in I are small 
when expressed in c.g.s. units of intensity. Hence the percentages may 
in some cases — especially with the lower fields — be several units in error 
for the less magnetic alloys. With the 30 and 48 per cent, alloys nothing 
* J. G. Gray and A. D. Ross, Proc. Roy. Soc. Edin ., xxix., 182 (1909). 
