repetition of M, Arago's experiments on magnetism^ ^c, 487 
28. Conceive no'w a plate of any thickness, and of indefinite 
superficial extent, of a metal or other magnetic, whose 
retentive power is very small. If either pole (suppose the 
north) of a magnet be brought vertically over a point in its 
surface, it will there produce a pole of the contrary name in 
the plate, the maximum of polarity being immediately under 
the magnet. Now let the magnet be moved horizontally along 
the surface, preserving the same distance from it. The points 
over which in succession it becomes vertical, not instantly 
receiving all the magnetism of which they are susceptible, 
will not have reached their maximum of polarity at the pre- 
cise moment of nearest appulse, but will continue to receive 
fresh accessions during the whole of that certain small por- 
tion of time when the distance (being at or near its minimum) 
undergoes no change, or only a certain very minute one. In 
like manner, the points which have attained their maximum 
of polarity, being left behind by the magnet, will by degrees 
lose their magnetism ; but the loss not being sudden, they will 
continue near their maximum for a certain finite time, during 
the whole of which the magnet continues receding from them, 
and leaving them farther and farther behind. Thus from 
both causes, there will be always in arrear of the magnet a 
space both more extensive and more strongly impregnated 
with the opposite polarity, than in advance of it ; and as the 
magnet moves forward, the point of actual maximum (or the 
pole ) of the plate, instead of keeping pace with it and being 
always precisely under it, will lag behind. There will thus 
arise an oblique action between the pole of the magnet and 
the opposite pole of the plate so lagging behind it ; and were 
the plate free to move in its own plane, the resolved portion 
