30 Voltaic. Induction. 
dynamic cylinder, or in right lines diverging from the magnetic axis 
NS. Let S denote a piece of steel placed within their influence. 
Now it is evident that were they to pass uninterruptedly, i. e. by the 
Fig. 3. 
xv 
lines y x, they would act with equal intensity upon the opposite sides 
of the line of particles a,a, a,a, &c., through which they pass, and 
thus destroy each others influence. But if we suppose that, upon 
entering the surface c, d, of the steel, they suffer a deflection from 
the magnetic axis NS, their action would be thrown into the dotted 
lines, v, r, and this would bring them in contact with only one side of 
every particle. Induction, therefore would commence upon the un- 
der sides of all the particles, a, above the magnetic axis NS, and 
upon the upper surfaces of those below it; the north poles of all the 
superficial particles, here only represented, would be repelled and 
revolve from N, while corresponding south polar forces would, 
in consequence of attraction, revolve around the particles, towards 
N ; and as the latter are those that zssue from the surface c d, they 
would give to the steel, S a south polarity, the opposite of that pos- 
sessed by N. The duration of this rotation, after the magnet is with- 
drawn, most probably depends upon the molecular condition of the 
steel, since we find that caloric, which expands the particles, enables 
the forces to return toa state of neutrality.* 
* This mode of explaining magnetic induction, is offered as mere conjecture. It 
does not, indeed, account for the absence of a voltaic current; but is not opposed to 
the supposition that the circulation is confined to the particles, individually. 
e 
