May 3, 1894] 



NA TURE 



Currents. 



Everyone knows that bodies can be divided into two 

 classes : conductors where we prove the transference of 

 electricity, that is to say, of voltaic currents, and in- 

 sulators or dielectrics. To the old electricians dielectrics 

 were purely inert, and their part consisted in opposing 

 the passage of electricity. If this were so, we could re- 

 place any insulating body by another of a different kind 

 without changing the phenomena. Faraday's experi- 

 ments have shown that it is nothing of the kind. Two 

 condensers of the same shape and dimensions put in 

 communication with the same sources of electricity will 

 not take the same charge (even if the thickness of the 

 isolating wire be the same), if the fta/itre of the isolating 

 [matter differs. Maxwell had made too deep a study of 

 I Faraday's works not to understand the importance of 

 dielectric bodies and the necessity of restoring to them 

 their proper function. 



Besides, if it be true that light is but an electric pheno- 

 menon, it follows that when it is propagated through an 

 insulating body, this body is the place of the phenomenon, 

 therefore there must be electric phenomena localised in 

 dielectrics ; but of what nature are they .- Maxwell 

 answers daringly : they are currents. 



All the experiments up to his time seemed to contradict 

 Ithis ; currents had never been observed except in con- 

 jductors. How could Maxwell reconcile his audacious 

 {hypothesis with such a well-founded fact ? Why do the 

 jhypothetical currents under certain circumstances pro- 

 Iduce manifest effects, which under ordinary conditions 

 !remain absolutely unobservable .■■ 



! It is because dielectrics oppose to the passage of 

 |electricity, not a greater resistance than the conductors, 

 !but a resistance of a different kind. A comparison will 

 make Maxwell's thought clearer. 



If we endeavour to bend a spring, a resistance is 

 ^encountered which increases in proportion as the spring is 

 ibent. If, therefore, we have at our disposal only a limited 

 force, a moment will come when the resistance being un- 

 surmountable, the movement will stop and equilibrium be 

 established ; at last, when the force ceases to act the 

 spring will bound bark, giving back all the work 

 expended to bend it. 



.Suppose, on the contrary, that we wish to move a 

 body immersed in water. Here again we meet with 

 jresistance which will depend on the velocity, but which, 

 if this velocity remains constant, will not increase in pro- 

 portion as the body advances ; the movement will 

 therefore continue as long as the force acts, and 

 equilibrium will never be attained ; finally, when the 

 [force ceases to act,thc body will not tend to return, and the 

 energy used for making it advance cannot be restored ; it 

 fvill have been entirely transformed into heat by the 

 jviscosity of the water. 



I The contrast is manifest, and it is necessary to dis- 

 tinguish between I'/cisiic and 7/isi-ous resistance. Then 

 dielectrics would behave, for electric movements, like 

 jslastic solids in the case of material movements, whilst 

 ponductors would behave like viscous liquids. Hence two 

 rategories of currents : current of displacement or Max- 

 well's currents which traverse dielectrics, and the ordinary 

 :onducting currents which circulate in conductors. 



The first, having to overcome a sort of elastic resis- 

 ;ance, can be but of short duration ; for, this resistance 

 ncreasing continually, equilibrium will be rapidly 

 established. 



The currents of conduction, on the contrary, having to 

 jvercome a sort of viscous resistance, can consequently 

 ast as long as the electromotive force which causes 

 hem. Let us lock again at the convenient comparison 

 •yhich M. Cornu has borrowed from hydraulics. 

 [Suppose we have water under pressure in a reservoir : 

 et us put this reservoir in communication with a vertical 

 ube ; the water will rise in it, but the movement will stop 



NO. 1279. VOL. 50] 



so soon as the hydrostatic equilibrium is reached. If 

 the tube is large, there will not be any friction, or loss 

 of charge, and water thus raised could be used for pro- 

 ducing work. We have here a picture of displacing 

 currents. 



If, on the contrary, the water of the reservoir flows out 

 by a horizontal tube, the movement will continue so long 

 as the reservoir is not empty ; but if the tube is narrow, 

 there will be a considerable loss of work, and a produc- 

 tion of heat by friction. We have here a picture of 

 conducting currents. 



Although it is impossible and of little use to try to re- 

 present to ourselves all the details of this mechanism, one 

 may say that all happens as if the displacement currents 

 had a number of little springs to bend. When the cur- 

 rents stop electrostatic equilibrium is established, and 

 the springs are so much the more bent as the electric 

 field is more intense. The work accumulated in these 

 springs, that is to say, the electrostatic energy, can be 

 wholly restored so soon as they can unbend themselves. 

 It is thus that mechanical work is obtained when the 

 conductors are allowed to obey the electrostatic attrac- 

 tions. These attractions would thus be due to the 

 pressure exercised on the conductors by the bent springs. 

 Finally, to follow the comparison to the end, the dis- 

 ruptive discharge must be likened to the rupture of over- 

 strained springs. 



On the other hand, the work employed for producing 

 conduction currents is lost and wholly transformed into 

 heat like that expended in overcoming the friction or 

 the viscosity of fluids. // is for this reason that the 

 cojiductim^ wires get hot. From Maxwell's point of view 

 there are only closed currents. Yox the old electricians 

 this was not so ; they looked upon a current as closed 

 which circulates in a wire joining the two poles of a 

 battery. But if, instead of reuniting the two poles 

 directly, one puts them in communication respectively 

 with the two armatures of a condenser, the instantaneous 

 current, which lasts until the condenser is filled, was con- 

 sidered open ; it went, it v.-as thought, from one armature 

 to the other across the wire of communication and the 

 battery, and stopped at the surface of the two armatures. 

 On the other hand, Maxwell supposed that the current 

 traverses the insulating plate, whioJi separates the two 

 armatures, under the form of a displacement current, and 

 that it is thus completely closed. The elastic resistance 

 which it meets on the passage explains its short duration. 



Currents can manifest themselves in three ways : 

 by their calorific effects, by their action on magnets and 

 currents, by the induced currents to which they give rise. 

 We have already seen why conduction currents develop 

 heat, and why displacement currents do not do so. On the 

 other hand, however, according to Maxwells hypothesis, 

 the currents which lie imagines, must, like* the ordinary 

 currents, produce electromagnetic, electrodynamic, and 

 inductive eftects. 



Why have we hitherto been unable to put these 

 effects in evidence ? It is because a displacement current, 

 however feeble, cannot last long, in the same direction ; 

 for the tension of our springs, ever increasing, would soon 

 stop it. There cannot therefore be in dielectrics, either 

 continuous currents of long duration, or sensible alter- 

 nating currents of long period. The effects will, however, 

 become observable if the alternation is very rapid. 



The Nature of Light. 



According to Maxwell, this is the origin of light. A 

 luminous ray is a series of alternating currents produced 

 in dielectrics, or even in the air or the interplanetary 

 vacuum, which changes its direction a thousand 

 billion times every second. The enormous induction due 

 to these frequent alternations produces other currents in 

 the neighbouring parts of the dielectric, and it is thus 

 that the luminous waves spread from point to point. 



