oncerning the making of an instrument for finding declination, the causes and manner of declination, and the different degrees of rotation in different places, the inclination of the stone, and concerning an instrument indicating by the influence of a stone the degree of declination from any horizon we have already spoken. Then we spoke about needles on the meridian of a stone, and their rotation shown for various latitudes by their rise toward the perpendicular. We must now, however, treat more fully of the causes of the degree of that inclination. Whilst a loadstone and a magnetick iron wire are moved along a meridian from the aequator toward the pole, they rotate toward a round loadstone, as also toward the earth with a circular movement. On a right horizon (just as also on the æquinoctial of ​ the stone) the axis of the iron, which is its centre line, is a line parallel to the axis of the earth. When that axis reaches the pole, which is the centre of the axis, it stands in the same straight line with the axis of the earth. The same end of the iron which at the æquator looks south turns to the north. For it is not a motion of centre to centre, but a natural turning of a magnetick body to a magnetick body, and of the axis of the body to the axis; it is not in consequence of the attraction of the pole itself that the iron points to the earth's polar point. Under the æquator the magnetick needle remains in æquilibrio horizontally; but toward the pole on either side, in every latitude from the beginning of the first degree right up to the ninetieth, it dips. The magnetick needle does not, however, in proportion to any number of degrees or any arc of latitude fall below the horizon just that number of degrees or a similar arc, but a very different one: because this motion is not really a motion of declination, but is in reality a motion of rotation, and it observes an arc of rotation according to the arc of latitude. Therefore a magnetick body A, while it is advancing over the earth itself, or a little earth or terrella, from the æquinoctial G toward the pole B, rotates on its own centre, and halfway on the progress of its centre^[237] from the æquator to the pole B it is pointing toward the æquator at F, midway between the two poles. Much more quickly, therefore, must the versorium rotate than its centre advances, in order that by rotating it may face straight toward the point F. Wherefore the motion of this rotation is rapid in the first degrees from the æquator, namely, from A to L; but more tardy in the later degrees from L to B, when facing from the æquator at F to C. But if the declination were equal to the latitude (i.e., always just as many degrees from the horizon, as the centre of the versorium has receded from the æquator), then the magnetick needle would be following some potency and peculiar virtue of the centre, as if it ​ were a point operating by itself. But it pays regard to the whole, both its mass, and its outer limits; the forces of both uniting, as well of the magnetick versorium as of the earth. *

---

The page and line references given in these notes are in all cases first to the Latin edition of 1600, and secondly to the English edition of 1900.

236 ^  Page 196, line 15. Page 196, line 18. De proportione declinationis pro latitudinis ratione.—Gilbert here announces, and proceeds in the next seven pages to develop, the proposition that to each latitude there corresponds a constant dip to a particular number of degrees. If this were accurately so, then a traveller by merely measuring the dip would be able to ascertain, by calculation, by reference to tables, or by aid of some geometrical appliance, the latitude of the place. In this hope Gilbert fought to perfect the dipping-needle; and he also worked out, on pages 199 and 200, an empirical theory, and a diagram. This theory was still further developed by him, and given to Thomas Blundevile (see the Note to p. 240). Briggs of Gresham College, on Gilbert's suggestion, calculated a table of Dip and Latitude on this theory. It was found, however, that the observed facts deviated more or less widely from the theory. Kircher (Magnes, 1643, p. 368) gives a comparative table of the computed and observed values. Further discovery showed the method to be impracticable, and Gilbert's hope remained unfulfilled.

237 ^  Page 197, line 18. Page 197, line 21. progressionis centri.—Note Gilbert's precision of phrase.