June 14, 1883] 



NA TURE 



161 



retain its magnetism, and that its natural state would be zero, or 

 neutrality. The apparent disappearance of magnetism, however, 

 is here due to the extreme freedom of motion of its molecules 

 allowing them at once to follow the comparatively feeble direct- 

 ing force of the earth's magnetism. We can demonstrate this 

 by feebly magnetising a rod of soft iron held vertically, so that 

 its north pole is at the lower portion. Upon removing the in- 

 ducing magnet, or electromagnetic coil, we find that the rod 

 retains a powerful north polarity ; but if magnetised in a contrary 

 sense, then we have only traces of magnetism left upon the 

 w.thdrawal of the inducing cause. To succeed in this experi- 

 ment, as in all others where soft iron is mentioned, we should 

 use the best Swedish charcoal iron, thoroughly annealed at high 

 temperature. 



We find, again, that rods of steel or iron will lose far less 

 magnetism when vibrated in the magnetic flip, or vertically, 

 when their north poles are at the lowest extremity, than when 

 horizontal, or still less than when their poles are contrary to 

 those of the earth's field, and also that they will acquire their 

 maximum magnetism from a given exciting cause when held 

 veiticallyas described, and the molecules allowed greater fret dom 

 of motion to obey the directing influence by vibrations, torsion, 

 stre-s, or blows upon the iron. Any influence that would tend 

 to give greater freedom of motion, such as heat or mechanical 

 trepidations, gives a far higher magnetic force to the iron than 

 could be obtained without these aids. 



In order to render visible the effects of motion upon magnetism, 

 we may take two glass tubes, or ordinary phials, of any length 

 or diameter, say, 10 centimetres in length by 2 centimetres in 

 diameter. If we now put iron filings in these tubes, leaving 

 about one-third vacant, so as to allow complete freedom in the 

 filings when shaken, we find that each tube, when magnetised, 

 retains an equal amount of residual magnetism, and that this all 

 disappears upon slightly shaking the tube. We are thus imi- 

 tating the effects of vibration. But if in one of these tubes we 

 pour melted re-in (in fact, any slightly viscous liquid, such as 

 petroleum, suffices), we then render these filings more rigid, and 

 then we can no longer produce by shaking the disappearance of 

 its residual magnetism. In pouring in petroleum we have appa- 

 rently been introducing a strong coercitive force, but we know 

 that it can only have the mechanical effect of rendering the iron 

 filings less free to turn, and so comparatively rigid. If we de-ire 

 Co see the effect of torsion, we have only to shake the filings so 

 that when the tube is held horizontally the vacant space is above, 

 and rotate it slightly (but without shaking) about a horizontal 

 axis. Its remaining magnetism instantly disappears upon rota- 

 tion, although we evidently have not changed the longitudinal 

 position of its particles. A similar effect takes place upon a soft 

 iron rod, for if we magnetise it and observe its remaining mag- 

 netism, we find that upon giving a slight torsion to this wire i's 

 remaining magnetism instantly disappears — a similar effect to 

 that in the rotating tube of iron filing-. But if the iron is 

 rendered more rigid by hammering, or steel rendered hard and 

 rigid by tempering, torsions or vibrations have but little effect, 

 as in the case of the filings rendered rigid as above mentioned. 

 Thus we have no 1 m-er need of an assumed mysterious coercive 

 force to account for (he retention of magnetism, for once know- 

 ing the mechanical qualities of iron and steel and their degree 

 of molecular rigidity or hardness, we can at once predict their 

 retent:ve magnetic powers. 1 



Rotation of Inherent Polarised Molecules. — Torsion, as well as 

 mechanical vibrations, has, as we have seen, a powerful influence 

 in aiding the molecules to overcome their inertia, and thus aid 

 them to rotate in the direction of the inducing influence ; and 

 we may thus polan-e strongly a flat, soft iron rod by simply 

 bending or vibrating it when held vertically, and if we measure 

 the magnetic force obtained we shall notice that the force is 

 strictly relative to the degree of softness of the iron. Thus, 

 with hard steel we should obtain only traces of polarisation, 

 whilst with extremely pure, soft Swedish iron we obtain the 

 maximum of force. The bar of iron or steel, being held in the 

 earth's magne.ic field, of infinite size compared wiih the bar, 

 and infinitely homogeneous, cannot deflect or weaken its sur- 

 rounding field. Its lower portion, being north, apparently 

 strengthens it by its reaction, whilst its upper, south, apparently 

 weakens the field ; but, as Maxwell has shown, " the two poles 

 of each molecule are equal and opposite, consequently the sum 

 of each molecule and the whole mass must be zero." 



' "On the Molecular Rigidity of Tempered Steel," by ftof D. E. 

 Hughes, F.R.S. sFroceedings Institution ff/ " Mechanical Engineers, pp 

 72-79, January, 1883.) 



We have a far greater induced polarity in iron or steel when 

 the iron is in thin bars or small wires, and this we should expect, 

 as the external molecules rotate directly under the influence of 

 the earth's magnetism, whilst those forming the interior of the 

 bar either rota'e feebly, or, as in the rase of very thick bars, 

 actually act as an armature, preventing by their influence free 

 rotation of the exterior molecules. 



Thus, as the sum of the two and equal polarities in a bar of 

 iron is zero, it is evident that its polarity must -be inherent. I 

 have some remarkably pure soft Swedish iron wire, one milli- 

 metre in diameter, and as its inherent polar force seemed great 

 when held vertically in the earth's magnetic field, I measured in 

 the induction balance this force compared with a similar column 

 of the magnetic atm "tsphere which it displaced. The inherent 

 polarity of this wire, simply rendered evident by the earth's 

 magnetism, was 15,600 times greater than the column it 

 displaced. 



We cannot, either by induction, conduction, or concentration, 

 produce a greater force in another body of similar displacement 

 or size, otherwise we could easily create power from a feeble 

 source. Thus the enormously greater magneiic power observed 

 in iron than the same column of air which it displaces must be 

 due to the inherent polarity of its molecules. 



Among-t numerous bars of iron upon which I have experi- 

 mented, one of ordinary hoop-iron, 2 centimetres wide, 40 

 centimetres long, and I J millimetre thick, not softened, possesses 

 sufficient molecular rigidity to be apparently uninfluenced by the 

 earth's magnetism. When this rod is rendered neutral we have 

 but feeble polarity — mere traces w hen it is held vertically under 

 the earth's magnetic influence ; but if we apply a few successive 

 torsions or vibrations to it when thus held, we have at once 

 several thousand times greater polarity than before. Now, if 

 iron had the power of deflecting or concentrating the earth's 

 magnetism upon itself, it should not require the mechanical aid 

 to molecular rotation given to it by these torsions or vibrations. 

 Thus we are forced to conclude at least the existence of the 

 inherent polarity of the molecules ; rind, if we admit this, we 

 must also as a necessary consequence, admit the rotation of 

 these molecules, else we cannot explain why mechanical vibra- 

 tions allowing freedom of motion should always produce the 

 polarity in accordance with the directing cause. I have already 

 shown that torsion and vibrations per se are apparently destruc- 

 tive of magnetism ; consequently in this case Poisson's two 

 fluids and Ampere's parallel currents should, according to their 

 theory, be mixed or'heterogeneous, whilst, according to the views 

 I am sustaining, the polari ed molecules should obey, as compass 

 needles, any magnetic directing cause whenever sufficient mole- 

 cular freedom of motion allows free rotation. 



The inherent polarity of iron may again be observed by draw- 

 ing a flat rod of soft iron over one or both poles of a permanent 

 magnet. This rod will then be powerfully magnetised, its 

 remaining magnetism, when separated from the magnet, being 

 sufficiently powerful to strongly deflect a suspended direction 

 needle. A few slight torsions or vibrations will then com- 

 pletely di- charge it. Now, suppose this operation repeated 

 succe-sively many thousand limes, if there was 1.0 inherent 

 polarity we should have gradually drawn all the polarity out of 

 the magnet, and discharged it into the atmosphere. Nothing of 

 the kind takes place. The molecules of the iron are simply 

 rotated each time, and the only energy in work expended or lost 

 comes from the arm of the experimenter, and the energy required 

 would be strictly in accordance with the molecular freedom, or 

 softness and hardness of the iron and steel ; thus, whilst soft 

 iron could be easily polarised and discharged by mechanical 

 torsion-, hard tempered steel would require a far greater 

 amount. 



Dr. Warren de la Rue, F.R.S. , kindly aided me in this part 

 of the research by passing a current from his well-known chloride 

 of silver battery through iron and steel wires. A condenser of 

 42-8 microfarad capacity, charged by 3-360 cells, was used. We 

 passed this enormous electric charge longitudinally through the 

 wires, and observations were made as to whether any change 

 whatever was produced in their quality or inherent polarity, the 

 result being that these wires gave exactly the same magnetic 

 polarity from a given directing or inducing cause as before, 

 being similar in nature and degree, c msequently this en irmous 

 electric force had not changed or destroyed the oiiginal inherent 

 polarity. 



If the molecules possess inherent polarity and rotate upon 

 their axes, similar to a series of compass needles having a slight 



