The arc of variation is not changed equally _in proportion to the distance of places_.
In the open sea, when a vessel is borne by a favourable wind along the same parallel, if the variation be changed by one degree in the course of one hundred miles, the next hundred miles do not therefore lessen it by another degree; for the magnetick [needle] varies erratically as respects position, form, and vigour of the land, and also because of the distance. As, for example, when a course from the Scilly Isles to Newfoundland has proceeded so far that the compass is directed to the true pole, then, as the vessel proceeds, in the first part of the course the variation increases toward the north-west[222], but rather indistinctly and with small difference: thence, after an equal distance, the arc is increased in a greater proportion until the vessel is not far from the continent: for then it varies most of all. But before it touches actual land or enters port, then at a certain distance the arc is again slightly diminished. But if the vessel in its course should decline greatly from that parallel either toward the south or the north, the magnetick [needle] will vary more or less, according to the position of the land and the latitude * of the region. For (cæteris paribus) the greater the latitude the greater the variation.
An island in Ocean does not change the variation[223], as _neither do mines of loadstone_.
Islands, although they be more magnetick than the sea, yet do not change the magnetick directions or variations. For since direction is a motion derived from the power of the whole earth, not from the attraction of any hill but from the disposing and turning power of the whole; so variation (which is a perturbation of the direction) is an aberration of the real turning power arising from the great inequalities of the earth, in consequence of which it, of itself, slightly diverts movable magneticks toward those which are the largest and the more powerful. The cause now shown may suffice to explain that which some so wonder at about the Island of Elba (and although this is productive of loadstone, yet the versorium (or mariners' compass) makes no special inclination toward it whenever vessels approach it in the Tyrrhenian sea); and the following causes are also to be considered, viz.: that the virtue of smaller magnetick bodies extends scarcely or not at all of itself beyond their own mines: for variation does not occur because of attraction, as they would have it who have imagined magnetick poles. Besides, magnetick mines are only agnate to the true earth, not innate: hence the whole globe does not regard them, and magneticks are not borne to them, as is demonstrated by the diagram of eminences.