206 Mr. R. H. M. Bosanquet on 



it be said that it is a natural expression of such facts to assume 

 that the magnetism depends on the force and on the conduc- 

 tivity of the material of disk or sphere, leaving the existence 

 of the circuit out of the question ? It would be equally sen- 

 sible to take a piece of copper out of a voltaic circuit, and 

 found our theory of current electricity on the hypothesis that 

 an electromotive force acting on the copper produced a current 

 through it proportional to its conductivity and to the force, 

 irrespective of the completion of any circuit. 



There is in magnetism this circumstance, which gives some- 

 what more colour to the hypothesis than there would be in the 

 above case — namely, that space conducts magnetism much 

 better than it does electricity; and in consequence some mag- 

 netic induction can always be set up in iron by a magneto- 

 motive force — just as, if we lived in the sea, and had voltaic 

 batteries and dynamo machines there, the action of an electro- 

 motive force would always produce some current, the circuit 

 being completed by the conducting-power of the sea-water. 



The second objection is to the substitution of the so-called 

 " magnetizing force " for a magnetomotive force. This is just 

 as if, living in the sea, we associated electromotive forces with 

 the currents in the sea-water which would inseparably accom- 

 pany them, took these currents for the measures of the forces, 

 and called them the electrizing forces. 



In carrying out the ordinary theory on this basis, we have 

 to suppose that the magnetizing force <£j within a magnetic 

 body has the power of remaining separate and distinct from 

 the magnetic induction as a whole, though the two are quan- 

 tities of the same nature. This has always seemed to me to 

 present insuperable difficulties as a physical conception. 



So soon as we replace the " magnetizing force " by a differ- 

 ence of potential or magnetomotive force, we can assimilate 

 the whole conception to that of the origin of an electric cur- 

 rent under an electromotive force. The quantity «£) becomes 

 merely the magnetic induction in vacant space, and 23 that 

 in magnetic matter. 33 replaces «£>, and is not supposed to 

 include it as before. According to the ordinary theory, 



33 = ^ + 47r3 or //,= 1 + 4t™; 

 where 



_ 23 _ 3 cv _ moment 

 ^W K ~W J ~~ volume' 

 The change of conception and the real meaning of the formula 

 can be shown as follows: — 



Let /jl= 1 + X. Suppose an infinite* bar acted on by a mag- 



* The bar must be infinite, not merely long. Rowland found the influ- 

 ence of the ends still sensible in the longest bars he tried. 



