Magnetite seems to have acquired its polarity in the same manner, namely by the earth's induction, when the ore contains a large enough percentage of pure iron. A large specimen (6 in. long by 3½ deep and weighing 5½ lb.) which I obtained from near Pilot Knob, Missouri, exhibits polarity, not only at its lateral ends, but also vertically, as the lower surface attracts the unmarked end of a needle, while the plane, which evidently occupied the upper surface in its native bed, attracts the marked end of the needle.
Iron fences invariably exhibit only the polarity by vertical induction; so also small buckets, bells, etc. But in the case of a bell about 3 ft. in diameter at its base, and over two feet deep, tapering to about a foot in diameter at the top, I found that although the top attracted the marked end of the needle, the bottom attracted the unmarked end of the needle only around the northerly half of the circumference, while the southern portion of this lower rim attracted the marked end in consequence of lateral induction, as in N. and S. rails.
Thus, upon a comparison of all these facts, it would appear that, if the magnetism induced by the earth is due to so-called currents of electricity, those currents must be underneath the rails, and must move from west to east, under the south to north rails, and from south to north under the west to east laid rails, as indicated by the arrows in the diagram.
This accords perfectly with what we should theoretically expect, in our northern hemisphere, if the electricity in the earth's crust is due to thermo-electrical currents from east to west, namely, from the more heated to the less heated portion, on any given latitude, while the earth revolves from west to east; as well as also from electrical currents trending from tropical to Arctic regions.
As the network of iron rails spreads from year to year more extensively over our continent, it will be interesting to observe whether or not any effect is produced, meteorological, agricultural, etc., by this diffusion of magnetism.
It may further interest some of your readers to have attention called to facts indicating
SYNCHRONOUS SEISMOLOGY.
The year recently closed furnishes interesting corroborative testimony of an apparent law regarding the propagation of earthquake movements most readily along great circles of our globe, as well as evidence that these seismic movements are frequently transmitted along belts (approximating to great circles) coincident sometimes with continental trends, at other times with fissures which emanate in radii at every 30°, around the pole of the land hemisphere in Switzerland, as described in one of my papers, read at the Montreal meeting of the A.A.A.S.
The terms synchronism or synchronous, as here used, are not designed to imply absolute simultaneity (although that is sometimes the case with disturbances 180° apart), but are rather intended to indicate the tendency presented by these phenomena to exhibit this internal activity, during successive days, weeks, or even months, along a given great circle of the earth, especially one or more of those connected with the land center; perhaps most of all along the great circle which forms the prime vertical, when the center of land is placed at the zenith.
In order to test the above, let us examine the record of the most prominent earthquakes or volcanic eruptions for the year 1883.