in the sense that polarity changes when
Mr. Gary moves his piece of sheet iron with its attached shingle nail
across the pole or near the pole of a magnet." "The most delicate
instruments fail to detect such a change of polarity," etc. Mr. Gary's
claim of a neutral line is of course absurd, but you are wrong in
saying that the polarity does not change under the conditions
described in the _Harper's Monthly_ article. Mr. Gary is perfectly
correct in claiming a change of polarity in that experiment, although
his other claim of deriving from this change of polarity a continuous
motion without consuming energy are manifestly absurd.

[Illustration: Gary Motor A.]

[Illustration: Gary Motor B.]

The change of polarity is easily explained. If a bar of soft iron,
whose length is two or three times the distance between the poles of
the horseshoe magnet, be placed in front of the latter as in the
sketch, and at some distance, poles will be induced, as shown by the
letters N S. Now let the bar approach the magnet. When within a short
distance consequent points will be formed and the polarity at the ends
will be reversed, the bar having four poles, as in the second sketch.
The bar of soft iron must have certain dimensions depending on the
size and power of the horseshoe magnet. By using a powerful
electro-magnet in place of a permanent one, a soft iron bar of
considerable size may be used, and the change of polarity exhibited by
showing the repulsion in one case for the south pole and in the other
for the north pole of a heavy permanent magnet. When in the proper
position a very small movement of the soft iron bar is sufficient to
produce the change.

WM. A. ANTHONY.

Cornell University, Ithaca, N. Y., March 2, 1879.

* * * * *

GARY'S NEUTRAL LINE.

_To the Editor of the Scientific American:_

I have just read the article in the issue of March 8, on the Gary
Motor, and cannot refrain from offering a suggestion on the subject.
When I read the article referred to in _Harper's_, I formed the same
opinion of the so-called invention that the writer in the SCIENTIFIC
AMERICAN has expressed, and, in the main, such is my opinion still. I,
however, tried the experiment by which Gary claims to prove the
existence of his neutral line, and soon found the same explanation
that the writer in the AMERICAN has given. I then, curiously enough,
modified the experiment in precisely the manner he suggests