602 
PROFESSOR J. A. EWING ON EXPERIMENTAL 
Load. 
Magnetometer 
reading. 
3. 
kilos. 
Field raised to 3\33 .... 
0 
173 
144 
Began loading. 
1 
246 
205 
2 
341 
292 
3 
467 
389 
4 
586 
409 
5 
678 
565 
&c. 
At about 10 kilos, a maximum was passed. The load was increased up to 13 kilos., 
then removed, and the cycle 0 — 13 — 0 was repeated three times, with the result that 
the changes of magnetism became nearly cyclic. At the third repetition the following 
readings were taken :—• 
Load. 
Magnetometer. 
3. 
Load. 
Magnetometer. 
3. 
0 
682 
569 
13 
837 
697 
1 
690 
575 
12 
844 
703 
2 
711i 
593 
11 
853 
711 
3 
751 
626 
10 
863 
719 
4 
794 
662 
9 
873 
727 
5 
836 
697 
8 
883 
736 
6 
869 
724 
7 
892 
744 
7 
890i 
742 
6 
901 
751 
8 
899 
749 
5 
905 
754 
9 
897 
747 
4 
900 
750 
10 
887 
739 
3 
879 
732 
11 
873 
727 
2 
835 
696 
12 
856 
713 
1 
774 
645 
13 
837 
697 
0 
690 
575 
This cycle is shown in Plate 63, fig. 40, where it is marked IV. The dotted fine 
running up through the diagram shows the effect of the loads which were first applied 
after the magnetising force had been brought to the value 3'33, at which it was kept 
during the remainder of the operation. Here, as in previous experiments to which 
attention was drawn in § 81, the first process of loading finds the magnetism in an 
exceedingly sensitive state, ready to respond by leaps and bounds to the successive 
increments of stress. The conditions, however, are different in the two cases. The 
wire of § 81 had previously been annealed in the magnetic field. In the present 
instance the field had been gradually increased to a constant value before the 
load was applied. We know already that the induced magnetism so reached is 
exceedingly sensitive to mechanical vibration (cf. § 49), and we now find a similar 
sensitiveness to application of stress, although unaccompanied by any vibration. 
When the action is repeated often enough to bring the changes of magnetism to a 
