108 CORRELATION OF PHYSICAL FOKCES. 



matter, as lines of which the extremities repel each other in 

 a definite direction; thus, if a line A B represent a wire 

 affected by electricity, and superposed on c D a wire affected 

 by magnetism, the extreme points A and B will be repelled to 

 the farthest distances from the points c and D, and the line A 

 B be at right angles to the line c D ; and so, if the lines be 

 subdivided to any extent, each will have two extremities or 

 poles repulsive of those of the other. If the line of matter 

 affected by electricity be a liquid, and consequently have 

 "entire mobility of particles, a continuous movement will be 

 produced by magnetism, each particle successively tending, 

 as it were, to fly off at a tangent from the magnet : thus, 

 place a flat dish containing acidulated water on the poles of a 

 powerful magnet, immerse the terminals of a voltaic battery 

 in the liquid just above the magnetic poles, so that the lines 

 of electricity and of magnetism coincide ; the water will now 

 assume a movement at right angles to this line, flowing con- 

 tinously, as if blown by an equatorial wind, which may be 

 made east or west with reference to the magnetic poles by 

 altering the direction of the electrical current : a similar effect 

 may be produced with mercury. These cases afford an 

 additional argument to those previously mentioned of the 

 particles of matter being affected by the forces of electricity 

 and magnetism in a way irreconcilable with the fluid or 

 ethereal hypothesis. 



The representation of transverse direction by magnetism 

 and electricity appears to have led Coleridge to parallel it by 

 the transverse expansion of matter, or length and breadth, 

 though he injured the parallel by adding galvanism as depth : 

 whether a third force exists which may bear this relation to 

 electricity and magnetism is a question upon which we have 

 no evidence. 



The ratio which the attractive magnetic force produced 

 bears to the electric current producing it has been investigat- 

 ed by many experimentalists and mathematicians. The data 



