RESEARCHES IN' MAGNETISM. 
G01 
§ 85. The above operations are shown on the left hand side in fig. 39. The curves 
on the right hand side of the same figure refer to the following subsequent opera¬ 
tions. for which the observed numbers need not be given. After several repetitions 
of the cycle 0—13 — 0, still acting only on the residual magnetism, the curves ab were 
taken. Then the earth’s force was allowed to act (that is to say, the neutralising 
current in the solenoid was broken) and the cycle 0 — 13 — 0 was applied twice. The 
results of the second application are shown in the curves cd. Comparing cd with ab 
we see that in the inducing field the range of change of magnetism caused by change 
of stress is greater than when we are dealing with residual magnetism only, and the 
former changes occur of course at a higher position in the diagram than the latter ; 
but the general character of the curves is substantially the same in both cases. At 
the end of the cycle cd the magnetometer reading was 140, and on tapping the wire 
this fell to 105, showing how, as a result of hysteresis, the value of the magnetism for 
zero of load was abnormally raised. 
Next the neutralising current was again applied, and the wire was tapped 
(load = 0). The residual magnetism fell to 53. Loading and unloading then gave 
the curves ef which, in consequence of hysteresis, left the residual magnetism at the 
end greater by 18 scale divisions than it was (after tapping) at the beginning. Tapping 
again reduced it to very nearly its former value. 
Finally, another cycle of loads was applied with vibration before each reading, but 
the wire was broken at its bottom end by too violent tapping during the removal of 
load. The curves for this operation are g and h. Although during the operation 
the residual magnetism was suffering a great additional reduction by the combined 
effects of tapping and stress, a maximum of residual magnetism occurred in g at an 
intermediate value of the stress, as the curves taken without tapping lead us to 
expect. 
It appears, then, that the effects of stress on the magnetism of iron, annealed or 
drawn, are substantially the same, whether the magnetism be induced or residual. In 
the latter case the effects are complicated by the gradual working out of the residual 
magnetism by successive loadings, just as in the former case they are complicated by 
the gradual working in of the induced magnetism, but when any operation is repeated 
often enough to give nearly cyclic effects this complication is eliminated, and we then 
find the effects of stress to be substantially identical in the two cases. 
§ 86. Effects of Stress on Magnetism induced by Fields of Various Strengths (April 13, 
1882).—The iron wire of the last paragraph (diameter 0‘77 mm.), which had been 
broken close to one end, was again used, the length being now 30'5 centims. It was 
not re-annealed, but tested in the drawn condition to which it had been brought by 
the operations already described. By gradually increasing the current in the solenoid 
surrounding the wire, the magnetising field was brought to the value 3*33 c.g.s. units ; 
during this time there was no load on the wire. Loading was then begun, and every 
addition caused at first an enormous increase of magnetism. Thus :—- 
