We find, first, that the world may be divided into two unequal portions, over one of which the needle has a westerly, and over the other an easterly, declination. Along the boundary line, of course, the needle points due north. England is situated in the region of westerly magnets. This region includes all Europe, except the north-eastern parts of Russia; Turkey, Arabia, and the whole of Africa; the greater part of the Indian Ocean, and the western parts of Australia; nearly the whole of the Atlantic Ocean; Greenland, the eastern parts of Canada, and a small slice from the north-eastern part of Brazil. All these form one region of westerly declination; but, singularly enough, there lies in the very heart of the remaining and larger region of easterly magnets an oval space of a contrary character. This space includes the Japanese Islands, Manchouria, and the eastern parts of China. It is very noteworthy also, that in the westerly region the declination is much greater than in the easterly. Over the whole of Asia, for instance, the needle points almost due north. On the contrary, in the north of Greenland and of Baffin’s Bay, the magnetic needle points due west; while still further to the north (a little westerly), we find the needle pointing with its north end directly towards the south. In the presence of these peculiarities, it would be pleasant to speculate. We might imagine the existence of powerfully magnetic _veins_ in the earth’s solid mass, coercing the magnetic needle from a full obedience to the true polar summons. Or the comparative effects of oceans and of continents might be called into play. But unfortunately for all this, we have to reconcile views founded on _fixed_ relations presented by the earth with the process of _change_ indicated above. Let us consider the declination in England alone.