234 Lord Rayleigh on the Behaviour of Iron and Steel 



imposing a small force (under £H) . If this be in the opposite 

 direction to the previous large force, it will produce, in spite 

 of the compensating-coil, a very sensible effect ; for in this 

 case the movement from to — JH is in continuation of the 

 previous movement from 5H to 0. But subsequent appli- 

 cations and removals of ^H produce no visible effect upon the 

 needle, as would have happened from the first had the small 

 force operated in the positive direction. We may conclude, 

 then, that the compensation for small forces suitable when the 

 iron is nearly free from magnetization is not disturbed by the 

 presence of considerable residual magnetism. 



To examine the action of a small increment or decrement, 

 when the total force is relatively large, we must either intro- 

 duce a second magnetizing helix or effect the variation of 

 current otherwise than by breaking the circuit. I found it 

 most convenient simply to vary the resistance taken from the 

 box, so arranging matters that the small alteration of current 

 required could be effected by the insertion or removal of a 

 single plug. The corresponding change of current is obtained 

 by inspection of a table of reciprocals ; and it was readily 

 proved that within the admissible range of the apparatus the 

 compensation was just as effective whether a step (not ex- 

 ceeding iH) was made from zero or from a force (say) of 5H, 

 20 or 30 times as great as the increment or decrement itself. 

 It need scarcely be repeated that there is an exception as 

 regards the first step, in the case where it is in the same 

 direction as the large movement preceding it. 



At this stage the original magnetizing-coil, having been 

 arranged for the investigation of the smallest forces, was 

 replaced by another at a greater distance from the suspended 

 needle. When the magnetization of the iron in its various 

 parts fails to vary in strict proportion to the force, the effective 

 pole is liable to shift its position ; and this is an objection to the 

 horizontal arrangement adopted in the earlier experiments. 

 The helix was therefore placed vertically, the lower end of the 

 iron core being a trifle below the level of the magnetometer-' 

 needle. The upper pole was at such a distance as to give 

 but little relative effect. The length of the new helix, wound 

 like the other upon a glass tube, is about 30 centim. The 

 windings are in four layers, at the rate altogether of 65 per 

 centim. ; so that (under the same current) the magnetizing 

 force is about twice as great as before. The resistance is 

 4*75 ohms. 



A large number of observations have been made upon a 

 core of rather hard Swedish iron, 3*30 millim in diameter. 

 The same compensating-coil as before was found suitable, and 



