, placing
the magnet in a vertical position, and using first a piece of sheet
iron and then an iron wire under it. This was before seeing the
article in the SCIENTIFIC AMERICAN. My experiment is well illustrated
by the writer's diagram, except that the nail should be at the end of
the iron wire, where its polarity is of course most strongly marked.
But the result is not as he states it. For, as the wire is brought up
toward the magnet, the nail drops off before the wire touches the
magnet. When the sheet iron is used, the point at which the nail drops
off is farther from the magnet than in the case of the wire, and when
it is brought nearer it will again pick up the nail, which then
continues to cling until the iron touches the magnet and afterwards.
Thus the existence of a line in which the soft iron, or induced
magnet, does not attract the nail, and above and below which it does
attract it, is demonstrated. That the polarity of the induced magnet
is reversed when it crosses this line may be demonstrated as follows:
When it is held beyond (or below) this line (Fig. 1), the negative
pole of the permanent magnet, the positive being kept at a distance,
may be made to approach the iron and touch it, without causing the
nail to drop. (Fig. 3.) But when contact occurs, the whole of the iron
must possess the polarity of that part of the magnet which it touches,
namely, negative. Hence in the position indicated in Fig. 1, the
polarity of the induced magnet does not correspond with that of the
permanent magnet, but is as indicated by the letters. On the other
hand, if the positive pole alone be made to approach, the nail will
drop; but when it is very near, or in contact, it again holds the
nail, and the iron is now positive; and if the negative pole also be
now brought into contact, the polarity of the soft iron will
correspond with that of the magnet, as shown in Fig. 2.

[Illustration: Gary's Neutral Line A.]

[Illustration: Gary's Neutral Line B.]

[Illustration: Gary's Neutral Line C.]

These experiments should be performed with the soft iron under both
poles of the magnet, and the ends of the former should extend somewhat
beyond the poles of the latter, or the nail is liable to jump to the
magnet as the "neutral" line is crossed. The position of the letters
in Fig. 1, of the previous article, represents the polarity of the
induced magnet to be the same as that of the permanent, which is true
only within (or above) th