262 Mr. Knox on the Direction and Mode of 



circuit, whatever increases the chemical action upon one more than upon the 

 other, increasing the rapidity of alternate states of induction, produces a charge 

 in a shorter time ; and this takes place not only when two different metals are 

 employed, but also, when plates of the same metal being used, a difference of 

 polish or a difference of heat applied alters the chemical action upon one plate 

 more than upon the other. A further analogy is faintly borne out by the 

 following experiments, which may lead to an explanation of some curious facts 

 regarding the alternate increase and decrease of intensity in the voltaic pile, 

 dependent upon the number of alternations, as observed by Delarive* and others. 

 Having connected, by means of insulated copper wires, the insulated conductors 

 of an electrical machine, with two insulated brass balls, the spark that passed 

 between the two balls measured one-fourth of an inch. When the insulated 

 negative conductor of this machine was connected with the insulated prime 

 conductor of another similar one, and its insulated negative conductor with one 

 of the brass balls, and the two machines rotated simultaneously, the length of 

 the discharging spark was increased to one-half; with three electrical machines 

 similarly arranged, the length of the spark which passed was one-third ; with 

 four, it returned to one-half; beyond this number no regularity in the length 

 of the discharging sparks was observable. The quantity in the electrical ma- 

 chine increases with the number of collecting forks, when the rubbers and forks 

 are disposed in such a manner, that the latter can receive the greatest quantity 

 of electricity from the excited glass ; so in the voltaic pile, the quantity is as the 

 number of particles of hydrogen set free against the surface of the platinum. 



The effect which a current of electricity, considered to be a row of particles 

 whose oppositely electrified surfaces are ranged in the same direction, undergoing 

 alternate states of induction and equilibrium, produces upon contiguous particles, 

 should be to induce in them oppositely electrified surfaces, which, in undergoing 

 alternate states of induction and equilibrium, should obviously give rise to a cur- 

 rent of electricity in an opposite direction, — and this is agreeable to fact. 



To afford an explanation of magnetism, considered as an electrical phenomenon, 

 no theory as yet proposed is adequate. That of Ampere (although exceedingly 

 beautiful) is yet all but physically impossible, for how can we suppose that when 

 the electrical current which magnetizes a steel bar ceases, the electricity in the 



* Bib. Univer. Tom. iv. p. 360. 



