Gilbert made a “terella,” or miniature of the earth, as a magnet, and not only demonstrated how the compass needle sets along the lines joining the north and south magnetic poles, but explained the variation and the dip.  He imagined that the magnetic poles coincided with the geographical poles, but, as a matter of fact, they do not, and, moreover, they are slowly moving round the geographical poles, hence the declination of the needle, that is to say its angle of divergence from the true meridian or north and south line, is gradually changing.  The north magnetic pole of the earth was actually discovered by Sir John Ross north of British America, on the coast of Boothia (latitude 70 degrees 5’ N, longitude 96 degrees 46’ W), where, as foreseen, the needle entirely lost its directive property and stood upright, or, so to speak, on its head.  The south magnetic pole lies in the Prince Albert range of Victona Land, and was almost reached by Sir James Clark Ross.

The magnetism of the earth is such as might be produced by a powerful magnet inside, but its origin is unknown, although there is reason to believe that masses of lodestone or magnetic iron exist in the crust.  Coulomb found that not only iron, but all substances are more or less magnetic, and Faraday showed in 1845 that while some are attracted by a magnet others are repelled.  The former he called paramagnetic and the latter diamagnetic bodies.

The following is a list of these.—­

Paramagnetic     Diamagnetic
Iron             Bismuth
Nickel           Phosphorus
Cobalt           Antimony
Aluminium        Zinc
Manganese        Mercury
Chromium         Lead
Cerium           Silver
Titanium         Copper
Platinum         Water
Many ores and    Alcohol
salts of the     Tellurium
above metals     Selenium
Oxygen           Sulphur
Thallium
Hydrogen
Air

We have theories of magnetism that reduce it to a phenomenon of electricity, though we are ignorant of the real nature of both.  If we take a thin bar magnet and break it in two, we find that we have now two shorter magnets, each with its “north” and “south” poles, that is to say, poles of the same kind as the south and north—­magnetic poles of the earth.  If we break each of these again, we get four smaller magnets, and we can repeat the process as often as we like.  It is supposed, therefore, that every atom of the bar is a little magnet in itself having its two opposite poles, and that in magnetising the bar we have merely partially turned all these atoms in one direction, that is to say, with their north poles pointing one way and their south poles the other way, as shown in figure 27.  The polarity of the bar only shows itself at the ends, where the molecular poles are, so to speak, free.