Prof. D. E. Hughes. 



[May 10, 



most rigid material — cast steel — had become free at red heat, and 

 rotated under the earth's magnetic influence, giving exactly the same 

 force on each, consequently the previous magnetisation of two of these 

 bars had neither augmented nor weakened the inherent polarity of 

 their molecules. Soft iron gave under these conditions by far the 

 greatest force, its inherent polarity being greater than that of steel. 



The gradual rotation of the molecules may be observed by strongly 

 magnetising a soft iron wire. If we then pass a feeble electric 

 current we find a slight diminution of magnetism after its passage, 

 and as we increase the force of the electric current the molecules almost 

 entirely rotate to zero ; and if we heat the wire to redness, or simply 

 put it in mechanical vibration, we have at last perfect neutrality, the 

 current having rotated the molecules as it does an exterior needle 

 perpendicular to itself, being simply Oersted's discovery of external 

 rotation of the needle applied to the molecule itself. The wire when 

 thus rendered neutral can, as Wiedemann remarked, again show 

 evidence of magnetism. The neutrality here is that known as chain 

 or circular magnetism, each molecule forming a link of a chain whose 

 circle of attractions is completed around the axis of the wire 

 through which the current has passed. 



In order to perceive the change of polarity which torsion produces 

 when an electric current is passing through an iron wire, the wire, which 

 maybe 2 milliras. diameter, and 40 centims. in length, should be placed 

 horizontally east and west, the centre of the wire being 3 or more 

 centims. above the axis of the needle, thus differing completely from the 

 position of the wire in Oersted's discovery, as in the latter case the 

 maximum rotation is obtained when the wire is parallel with the needle ; 

 but in the following experiments the maximum effect is obtained when 

 the needle is at right angles to the conducting wires, and none what- 

 ever if parallel, its action being perpendicular to the exterior rotations 

 of Oersted's discovery ; and a similar phenomenon will be observed in 

 all the rods mentioned later in the case of snperposed magnetism. 



An iron wire, under the conditions above mentioned, shows strong 

 north or south polarity by a left or right handed elastic torsion ; and 

 if we leave the wire with a slight remaining permanent torsion, 

 producing movement of the needle to the left, it diminishes upon 

 breaking the electric circuit, and increases to its previous value on 

 re-establishing the current ; this is due to a slight mechanical mole- 

 cular twist. If we now magnetise this rod slightly in a contrary 

 direction (the needle then having a movement to the right) we find 

 that upon re-establishing the current it now increases the right 

 deflection. Thus the slight magnetism has completely reversed the 

 influence of the previous mechanical torsion. 



The question here arises, have we rotated the molecules from their 

 previous position, or is it simply the reaction of magnetism considered 



