RESEARCHES IN MAGNETISM. 
607 
field was applied, and then a cycle of loads 1 — 6 — 1 was applied several times while 
the magnetic changes were noted. Except in the region of approximate saturation, 
the first effects of loading were positive, but in succeeding applications on gave 
diminution and off gave increase of magnetism, the curves for on and off forming a 
simple loop like that already described in speaking of the effects of stress on an 
annealed wire exposed to the vertical component of earth’s field. (§ 78.) 
The action was in this case examined in fields ranging up to 12 c.g.s. units, and 
the results are shown in Plate 64, fig. 43, where the curves are numbered as 
follows :— 
Curve. 
Magnetising field 
(in c.g.s. units.) 
I. 
0-68 
II. 
1-50 
III. 
2-05 
IV. 
273 
V. 
3-41 
VI. 
478 
VII. 
12-0 
The line which (in every case but curve VII.) slopes up to the right starts from the 
value of magnetism reached by applying the field to the (previously demagnetised) 
wire, and shows the effect of the first loading. The cycle of loads was not applied 
often enough to make the changes of magnetism cyclic, except at the highest mag¬ 
netisation, where (as in the stretched wire already referred to) a cyclic state is reached 
very quickly. In each field the effects of several successive loadings and unloadings 
are exhibited in fig. 43, in a way which inspection of the figure will make obvious. 
Curve la was obtained in the field 0'68 by tapping the wire and then applying and 
removing loads. 
Each kilo, of load corresponds in this wire to a stress of 2‘04 kilos, per sq. mm. 
§ 90. Curves of 3 cmcl <§ taken under Constant Load .—Concurrently with the series 
of experiments described in §§ 87-88, and represented in figs. 41 and 42, another series 
of observations was conducted on the same wires and with these in the same states as 
in the foregoing tests, by a wholly different method. Each wire, after being demag¬ 
netised by reversals, had a definite load applied to it. Then magnetising force was 
gradually applied, and the relation of 3 to <§ was observed. This gave a curve of 
magnetisation under that particular load. Then, after demagnetisation, a different 
load was put on, another curve of magnetisation taken by again applying magnetising 
force, and so on. The magnetising force was applied gradually and continuously by 
means of the slide described in § 18, and the magnetisation was observed by the 
direct magnetometric method. In fact, as regards measurement of magnetism and 
field, the conditions of each experiment were identical with those under which the 
changes of magnetism were caused by changing loads in a constant field, and the 
4 i 2 
