'JO 
MESSRS. G. r. C. SEAELE AND T. G. BEDFORD 
where m is a function of Hq. To determine m in terms of Hq we plotted m against 
Hq, and found that m is nearly proportional to as the following table shows ;— 
Ho. . . . 
4 
.0 
i 
10 
15 
m . 
•71 
•785 
•99 
1-15 
1-28 
•35Hh .• . 
•70 
•78 
•925 
1-10 
1-35 
Hence, with fair accuracy, the formula 
W = - GOO 
represents W as a function of Hq and Bq, over a large range, provided that the load 
producing torsion exceeds 100 grammes. The hooked jDart of each curve corresponds 
to small torsions due to loads of 100 grammes and under. The value of W for zero 
strain, recorded in § 57, is given with considerable accuracy by W = ’326 H’qBq — 320 
over a considerable region in the neighbourhood of the maximum permeabihty. 
§ 66. The second example is furnished by the exjDeriments on the torsion of a steel 
rod, described in § 55. The curves connecting W and Bq are shown in fig. 8 along 
with the W — Bq curve for zero stress. We again find that for the larger values of 
Hq those portion of the curves which correspond to the larger stresses are straight 
lines radiating from a single point—in this case the origin—and thus obtain 
W = wiBq. The values of m and Hq are given in the table. 
. 
10 
15-72 
26-64 
37-4 
m . 
1-25 
1-98 
2-94 
3-32 
•64(Ho-6-2p . . 
1-25 
1-97 
2-90 
3-57 
Thus, approximately, 
W=:-64(HQ-6-2)iBQ. 
This expression, would naturally fail to represent facts when Hq < 6'2. 
The value of W for zero stress, recorded in § 54, is given closely over the whole 
range by W = •57H"qBq —1800. 
§ 67. We now take the experiments on the effect of tension upon a soft iron wire 
described in § 52. The curves connecting W with Bq are shown in fig. 17 along with 
the W — Bq curve for zero stress. Each of the curves, which shew the effects of stress, 
is again made up of a straight part and a hook, the straight parts radiating from the 
point Bq = —600, W = —950. For the slope of the lines we have 
