2 Professor D. E. Hughes [Feb. 29, 



molecule, tlie molecule itself would then be electro-magnetic, and the 

 question still remains. What is polarity ? Have the supposed electric 

 currents separated the two assumed magnetic fluids contained in the 

 molecule, as in Poisson's theory ? or are the electric currents them- 

 selves magnetic, independent of the iron molecule ? 



In order to produce the supposed heterogeneous arrangement of 

 neutrality, Ampere's currents would have either to change their 

 position upon the molecule, and have no fixed axis of rotation, or else 

 the molecule, with its currents and polarities, would rotate, and thus 

 be acting in accordance with the theory of De la Kive. 3rd. This 

 theory does not explain why (as in the case of soft iron) polarity 

 should disappear whenever the exciting cause is removed, as in the 

 case of transient magnetisation. It would thus require a coercive force 

 in iron to cause exactly one-half of the molecules to instantly reverse 

 their direction, in order to pass from apparent external polarity to 

 that of neutrality. 



The influence of mechanical vibrations and stress upon iron in 

 facilitating or discharging its magnetism, as proved by Matteucci, 

 1847, in addition to the discovery by Page, 1837, of a molecular 

 movement taking place in iron during its magnetisation, producing 

 audible sounds, and the discovery by Dr. Joule, 1842, of the elonga- 

 tion of iron when magnetised, followed by the discoveries of 

 Guillemin, that an iron bar bent by a weight at its extremity would 

 become straight when magnetised ; also that magnetism would tend to 

 take off twists or mechanical strains of all kinds — together with the 

 researches of Matteucci, Marianini, De la Rive, Sir W. Grove, Faraday, 

 Weber, Wiedemann, Du Moncel, and a host of experimenters, in- 

 cluding numerous published researches by myself — all tend to show 

 that a mechanical action takes place whenever a bar of iron is 

 magnetised, and that the combined researches demonstrate that the 

 movement is that of molecular rotation. 



De la Rive was the first to perceive this, and his theory, like those 

 of Weber, Wiedemann, Maxwell, and others, is based upon molecular 

 rotation. Their theories, however, were made upon insufficient data, 

 and have proved to be wrong as to the assumed state of neutrality, 

 and right only where the experimental data clearly demonstrated 

 rotation. 



I believe that a true theory of magnetism should admit of 

 complete demonstration, that it should present no anomalies, and that 

 all the known effects should at once be exj^lained by it. 



From numerous researches I have gradually formed a theory of 

 magnetism entirely based upon experimental results, and these have 

 led me to the following conclusions : — 



1. That each molecule of a piece of iron, as well as the atoms of 

 all matter, solid, liquid, gaseous, and the ether itself, is a separate and 

 independent magnet, having its two poles and distribution of magnetic 

 polarity exactly the same as its total evident magnetism when noticed 

 upon a steel bar-magnet. 



