1S6 



KNOWLEDGE. 



[August 1, 1900. 



Two circular boards connected by elastic walls form 

 a diiun which can be filled with liquid. The upper 

 board is then hung from a horizontal whirling table, 

 while a weight is suspended from the lower one. When 

 the drum is spun round, the sides bulge out, and the 

 ends approach each other, raising the weight. A 

 magnetic field will be represented by means of a number 

 of chains, each made by attaching drums end to end, 

 and following the contour of a line of force in the field. 

 It will be seen then that when rotation is set up the 

 end bouudaries will be drawn together, representing 

 magnetic attraction, while the lines of force drive each 

 other apart sideways, representing magnetic repulsion. 



Professor Lodge indicates the whirls in an insulating 

 medium by means of cogwheels geai-ing into one another 

 and also into those of the conductor, and in order to 

 get over the difiiculty that two contiguous wheels must 

 be rotating in opposite directions, he assumes them to 

 be equivalent to positive and negative electricity 

 alternately. One of these models, representing a section 

 of a magnetic field, is illustrated in Fig. 5, the wheels 

 representing positive electricity being marked +, and 

 those representing negative electricity being marked — . 



If these rotate alternately in opposite directions 

 the electrical rotation or circulation in the field will 

 be all in one direction. In a medium of this kind, wi% 

 all the wheel work revolving properly, there will be 

 nothing of the nature of an electric current, for at every 

 point of contact of two wheels positive and negative 

 electricity respectively are travelling at the same rate 

 in the same direction, but a current may evidently be 

 represented by making the wheels gear imperfectly and 

 work with slip, and a line of slip among the wheels will 

 represent a linear current. 



Professor Lodge points out that such a line of slip 

 must always form a closed curve, as is required by the 

 fact that electricity must flow in a closed circuit. For 

 if only one wheel slip, the current coincides with its 

 circumference; if a row slip, the direct and return 

 circuits are on opposite sides of the row; and if an 

 area of any shape with no slip inside it is enclosed by a 

 line of slip the circuit may be of any shape but always 

 closed. 



In an insulator or dielectric there is no slip in the 

 gearing, so a conduction current is impossible, but a 

 metallic conductor must be considered as a case of 

 friction gearing with more or less lubrication and slip; 

 thus, turning one wheel will only start the next one 

 gradually, so that, until all the wheels ai-e in full spin, 

 there is a momentary current. In a perfect conductor 

 there must be no gearing, and such faultless lubrication 

 that no spin can be transmitted from one wheel to 

 another. 



In a magnetic medium, which is not magnetised, the 

 whirls are to be considered as taking place about axes 

 pointing indiscriminately in all directions, or, more 

 accurately, according to the researches of Professor 

 Hughes, the various chains of whirls must form closea 

 curves within the magnetic substance. 



When the medium is magnetised these are broken up, 

 and a preponderating orientation in a certain direction 

 takes place, and this may be most simply treated by 

 assuming that a certain proportion of the whirls are 

 accurately faced in this direction, the others facing 

 equally in all directions. 



When a magnetic disturbance is propagated through 

 an insulator in which all the wheels gear perfectly into 

 each other, propagation of spin through the mass will 

 take place with extreme rapidity, as there can be no 



slip, but only a slight distortion and recovery. In a 

 conductor, on the other hand, so long as the spin is 

 either increasing or decreasing, slip will be going on 

 throughout, and a certain time will elapse before a 

 steady state is attained. In highly magnetic sub- 

 stances, such as iron, and in a lesser degree nickel and 

 cobalt, we know that this time is greatly increased, 

 and may be represented in our model by increasing the 

 mass of the moving wheelwork, either by giving greater 

 mass to each of the wheels or by taking more of them, 

 or bv a combination of the two methods. 



Fig. 4. — Lodge's Hvclraulic \i 11 -I Lejden Jar. 



From Ludge's " Modern Views of Electricity.'' 



Take the case of a current in a copper wire gradually 

 increasing and producing magnetic spin in the surround- 

 ing medium. A section of the field through the wire 

 may be represented by a rack gearing into a train of 

 wheelwork, as shown in Fig. 6. As soon as the rack 

 begins to move the wheels will begin to rotate until 

 the whole of the surrounding medium is in a whirling 

 condition. Previous to a steady state of spin being 

 attained the motion of the rack will be opposed by the 

 inertia of the wheelwork, representing the opposing 

 E.M.F. of self-induction, or electro-magnetic inertia, and, 

 when the medium is in a state of spin, the stopping of 

 the rack will be opposed in a similar manner. If the 

 diagram is rotated round the rack the wheels become 

 circular vortex rings. As the distance from the rack 

 increases their cores increase in diameter, and therefore 

 the rate of spin diminishes, until at great enough dis- 

 tances the medium will hardly be disturbed. Slip 

 takes place entirely along the wire, while the axes of 

 spin are at right angles to it. If slip could take 

 place without friction, and the consequent dissipation 

 of energy in the form of heat, we should have the 

 analogue of a perfect conductor, if such a substance 

 existed. As a matter of fact no such substance is 

 known, and, therefore, in order to maintain a current in 

 a conductor, the energy continually being dissipated in 

 the form of heat must be continually supplied from some 

 source of power, such as a dynamo or battery. 



