476 Mr, Babbage and Mr. Herschel's account of the 
needle was made to deviate or revolve by the rotation of a 
plate beneath it. The motion of the needle must of course 
be rendered irregular by the effects of its polarity, and sub- 
ject to periodical accelerations or retardations ; and it is ob- 
vious, that in the case of a very weak magnetic force in the 
plate it can never execute an entire revolution, but must 
oscillate backwards and forwards till reduced to rest by the 
friction and resistance of the air. It occurred to us, how- 
ever, that much more regular and uniform results might be 
obtained by this means, could the polarity of the needle be 
destroyed without at the same time destroying its magnetism ; 
in other words, could the earth's action on it be so precisely 
neutralised as to allow of its resting indifferently in all direc- 
tions. The obvious mode of doing this, by the approach of 
a powerful magnet acting in opposition to the earth, proved 
much too coarse for our purpose, which however, after a few 
trials, we found might be accomplished to any required degree 
of precision by the following simple contrivance. 
If two exactly equal and similar magnets of equal strength 
be placed parallel to each other, but in a reverse position, and 
at such a distance as not mutually to affect each others' mag- 
netism, and if in this situation they be firmly attached to a 
piece of wood, glass, &c. the system so formed will have no 
polarity, i. e. no tendency to rest in one rather than another 
situation, however suspended. This is clear ; because what- 
ever be the inclination (6) of one of the magnets to the line 
of dip, that of the other will necessarily be (180 6), and 
the directive forces being represented by the sines of these 
two angles will always be equal and opposite, so that each 
magnet urges the system with equal force, but in opposite 
