It was, however, well known from the thrilling experience of Columbus and the constant report of travelers that this was not the case. Gilbert himself says that at the time of writing, in the year 1600, the needle pointed 11-1/3° east of north in London; but what he did not know and could not have known was that this easterly deviation was decreasing from year to year, to vanish altogether in 1657, after which the needle began to decline to the west. This magnetic declination sorely perplexed Gilbert, as it did not fit in with his theory. Yet an explanation was needed; and as the earth must be considered a normal and well-behaved magnet, though of cosmical size, Gilbert turns the difficulty by saying that this variation is nothing else than "a sort of perturbation of the directive force" caused by inequalities in the earth's surface by continents and mountain masses: "Since the earth's surface is diversified by elevations of land and depths of seas, great continental lands, oceans and seas differing in every way while the power that produces all magnetic movements comes from the constant magnetic earth-substance which is strongest in the most massive continent and not where the surface is water or fluid or unsettled, it follows that toward a massive body of land or continent rising to some height in any meridian, there is a measurable magnetic leaning from the true pole toward the east or the west." So convinced is Gilbert of the true and satisfactory character of his explanation that he goes on to say that, "In northern regions, the compass varies because of the northern eminences; in southern regions, because of southern eminences. On the equator, if the eminences on both sides were equal, there would be no variation." In a later chapter of Book IV., he adds that, "in the heart of great continents there is no variation; so, too, in the midst of great seas." As continents and mountain-chains are among the permanent features of our planet, Gilbert concluded that the misdirection of the needle was likewise permanent or constant at any given place, a conclusion which observations made after Gilbert's time showed to be incorrect. Gilbert writes: "As the needle hath ever inclined toward the east or toward the west, so even now does the arc of variation continue to be the same in whatever place or region, be it sea or continent; so, too, will it be forever unchanging."