1910-11.] On Magnetism of Copper-Manganese-Tin Alloys. 89 
Table II. — The Alloys as Cast. 
Test. 
Alloy. 
14 per cent, 
tin. 
16 per cent, 
tin. 
18 per cent, 
tin. 
30 per cent, 
tin. 
38 per cent, 
tin. 
48 per cent, 
tin. 
ho 
17 
20 
21 
? 
21 
? 
ho 
35 
43 
45 
? 
53 
0-4 
boo 
54 
61 
66 
? 
95 
1-0 
1 200 
70 
80 
80 
0*7 
159 
1-9 
hoo 
78 
88 
88 
TO 
208 
2-8 
I 
-*-r 
22 
13 
15 
? 
11 
0-6 
C.F. 
24 
10 
14 
? 
9 
ca. 50 
The values obtained for the different alloys are set out in six vertical 
columns. The first five horizontal lines give the intensities of magnetisa- 
tion, I 20 , . . . ., I 300 , in c.g.s. units corresponding to effective magnetising fields 
of 20, 50, 100, 200, and 300 gauss. In calculating these values allowance 
has been made for the self-demagnetising action of the specimens themselves 
by employing the factors investigated by Du Bois for iron cylinders.* 
These corrections have been shown to give satisfactory results when em- 
ployed in tests on copper-manganese-aluminium alloys.-]* They will prob- 
ably apply with like accuracy in the present instance, but in any case the 
errors introduced in the absolute values given must be small, and quite in- 
significant in the case of the relative effects which are here of chief import- 
ance. The sixth horizontal line in Table II. gives the intensity of the 
residual magnetisation, I,., when the specimen is acted on by zero force, that 
is to say, when the current in the magnetising solenoid is adjusted so as to 
give a magnetic field equal and opposite to that due to the magnetisation 
of the specimen. The seventh line exhibits the magnitude of the coercive 
force, C.F., in c.g.s. units. As this quantity and the residual magnetisation 
depend to a considerable extent on the maximum field employed in testing, 
they are taken here and in the subsequent tables to be the values obtained 
in a cyclic variation of the true effective field between the limits ±200 
gauss. It will be seen that the saturation values of I vary somewhat 
irregularly with the composition. Thus we have evidence of two groups 
of magnetic alloys, one group containing the alloys with 14-18 per cent, 
tin and the other group having the 30 per cent, alloy as a member. The 
magnetism of the 30 per cent, and 48 per cent, tin alloys is so feeble that 
the values given in the table must be taken merely as approximations. 
* H. du Bois, Ann. d. Phys. u. Chem., xlvi, 485 (1892). 
t A. D. Ross, Proc. Roy. Soc. Edin ., xxvii., 88 (1907). 
