August 1, 1900.] 



KNOWLEDGE 



185 



water representing the earth. In this model tlie gradual 

 distension of tlie indiarubber bag represents the charging 

 of the jar. In Fig. 4, two extra stopoocks. A' and W. 

 leading direct to the tank, have been added, to save the 

 trouble of disconnecting the pump in order to connect 

 A directlv with the tank, when illustrating the charging 

 of a jai- by alternate contact. 



If the two tubes shown in Fig. 2 are placed together 

 so as to form a single ring with two diaphragms across 

 it, both in a state of strain, then if one of these is 



1. 2. 



Fig. 1. — llviiraulio Model of a Circuit coosistiny partlj of Con- 

 ductors aud partly of lusulators. 



Fig. 2. — HvdrauHc Model of a pair of eq^ually and oppositely 

 charged Conductors. 



broken the pressures on the opposite sides of the re- 

 maining one will become equalised, and therefore the 

 diaphragm will return to its normal condition of flatness. 

 Owing, however, to the elasticity of the material of the 

 diaphragm, there will be a slight oscillation on either 

 side before it permanently assumes its position of equili- 

 brium. As far back as 1853, Lord Kelvin showed 

 mathematically that in general the discharge of a 

 Leyden Jar was oscillatory, and in 1859 and subsequent 

 years Feddersen confirmed this experimentally. 



A current of water cannot be started or stopped 

 suddenly, and similarly it is found that when an 

 electric current is started it takes an appreciable time, 

 though a very small one, to attain its full strength. 

 Again, when an electric current is arrested by breaking 

 the circuit, a very much larger spark is obtained than 

 the one observed on closing the circuit, and the more 

 sudden the break the larger the resulting spark. If, 

 however, an electric current really possesses inertia, as 

 a stream of water does, it should give rise to mechanical 

 as well as to electrical effects. These have been looked 

 for in vain by Clerk Maxwell, Professor Lodge, and 

 others. It maj' be that an electric current consists of 

 two equal streams in opposite directions ; or, rgain, 

 perhaps the hydraulic analogy is only of use in explain- 

 ing a few of tJie more obvious phenomena; and it 

 certainly does not account for the existence of the 

 magnetic fields in the neighbourhood of conductors 

 carrying electric currents or any of the phenomena 

 depending on them. 



The existence of these magnetic fields leads, moreover, 

 to the question whether we can regard the electricity 

 as forced along the conductor by a simple pressure, 

 analogous to that which drives water along a pipe, or 

 whether the energy required to maintain the flow is 

 transmitted through the insulating medium to every 

 portion of the boundary between it and the conductor. 

 Professor Poynting has shown that the latter is really 

 the case. The energy which drives an electromotor, for 

 instance, or maintains a series of electric lamps, is not 



convevcd through the conducting wires. The dynamo 

 gives its energy to the surrounding medium, thereby 

 inducing certain strains in it which spread in all 

 directions. If there were no conducting wires a per- 

 manent condition of strain would bo set up in the 

 medium, and when the energy reaches the conductors 

 some of it is dissipated, and the continuous How of a 

 cuiTent of electricity thus becomes possible. When an 

 attempt is made to transmit too much energy by means 

 of an electric cable it is the insulation and not the 

 copper wire which gives way. 



An electric current docs not start simultaneously at 

 every point in the section of a conductor, but tho 

 starting or stopping begins at the outside, and penetrates 

 inwards the more rapidly the worse the conductivity 

 of the material. If this were inrmitc the cunent would 

 never penetrate beyond the outer skin of the conductor. 



Professor Lodge illustrates this by the experiment of 

 spinning a tumbler of liquid, with some small particles 

 in suspension, to make the motions of the different 

 portions visible. The outer layers begin to move first, 

 and the motion gradually penetrates inwards, and when 

 the spinning of the tumbler is stopped the outside 

 portions of the liquid stop first. If the liquid is very 

 viscous, like treacle, the motions spread rapidly, 

 corresponding to a bad conductor of electricity, but if 

 extremely mobile then the inward propagation is much 

 slower, corresponding to a good conductor. The analogue 

 of a perfect conductor would be found in an absolutely 

 non-viscous liquid, and in such the motion would never 

 penetrate beyond the outermost skin. 



Suppose now we wind a conductor into a coil and pass 

 an electric current through it. We find that it behaves 

 in every way as a magnet, in fact it is a magnet as 

 long as the current continues to flow ; hence Ampere's 

 theory that magnetic substances owe their properties 

 simply to electric whirls in their molecules. These 

 whirls are not confined to the iron or steel of a magnet 

 but spread into the surrounding space, forming what is 



Fio. 3. — Skeleton Diagram of Lodge's Hydraulic Model of 

 Leyden Jar. 



From Lodge's " Modern Views of Electricity." 



known as the magnetic field, and this may be mapped 

 out by means of iron filings which cling, end to end, 

 along lines coinciding at every point with the direction 

 of the magnetic force, and are known a^ " lines of 

 force." These lines of iorco must constitute the axes 

 of molecular whirls, and every such line forms a closed 

 curve, part of which is in the iron, and the remainder 

 in the air or other surrounding media. The effect of 

 such whirls, if they consisted of a material fluid, may bo 

 illustrated by means of a model suggested by Professor 

 Lodge. 



