192 Prof. D. E. Hughes. 



[May 10, 



placement of the coil B necessary to find a zero. Thus, we can at 

 once find the slightest strain or fissure, such as partial rupture in iron 

 rods of any size or form. The coils may be close together or widely- 

 separated, and would find a practical application if applied to shafts 

 undergoing constant strain, such as the screw shaft of steamboats. 



The balance shown in fig. 3 is the one I generally employ, and is 

 preferable for the following experiments. , 



If we take a flat disk of iron similar in form to our usual current 

 coins, we find that if it is placed flat on or parallel with the coils, we 

 have a great reduction of induced currents upon the pair of coils in 

 which it is placed, due to the energy expended in creating the 



Arago " circular currents, and its action then is precisely similar to 

 copper or all non-magnetic metals, but if this disk is revolved 90°, or 

 placed perpendicular, it then acts simply as a magnetic body and 

 similar to the bar of iron F ; the induced currents on its pair of coils 

 are strengthened by the reaction of its electro -magnetic conductivity. 



That conduction itself is due to molecular rotation is proved by the 

 fact that soft iron shows a far higher conductivity than hard iron or 

 steel ; we observe here the same results of freedom and rigidity, and 

 all the previous effects of rotation are again repeated as conduction. 

 Soft Swedish charcoal iron shows such a marked superiority over all 

 other irons and steel, as regards its power or the force obtained, that I 

 feel convinced that the same superiority would be shown by its use for 

 the cores of all electro -magnets, particularly those of telegraph in- 

 struments, where rapid action and the maximum of force obtainable 

 from a feeble current are required. 



Faraday showed that iron loses its magnetic force at red-yellow 

 heat, and the induction balance is peculiarly adapted for investigating 

 this phenomenon. If we place an iron wire or rod, as at F, we find 

 ■on heating it that its conductivity gradually rises, being at black heat, 

 just before the visible red, double of that noticed at the ordinary 

 temperature, being a similar result to that previously noticed with the 

 single coil balance. But if we increase the heat until the rod becomes 

 red-yellow, all conductivity instantly vanishes and the iron apparently 

 has lost all its powers, being then similar to a piece of copper or other 

 non-magnetic metal; what takes place is, however, not destruction, 

 for, at red heat, its inherent polarity reappears instantly with its full 

 previous force. There seems no gradual diminution or reappearance 

 of its polarity, it is sudden and apparently instantaneous, its time of 

 action being less than the yoVo" P ar ^ °^ a secon( i. The induction 

 balance will, no doubt, in the future, enable me to investigate this 

 phenomenon. 



I find that the conducting power of soft iron is greatly reduced by 

 magnetising, generally one-fourth of its total conductivity, the 

 residual magnetism being in reality a partial rotation, thus reducing 



