On the Magnet/III-8
of a loadstone, and how they can agree and
stand joined together.
* end, and produce the appearance of a fork; and if one end be forcibly impelled toward the other, the other declines and bends away from association with it, as in the following figure.  upon the pole on account of their nearness to one another; either alone would otherwise stand erect and perpendicular. For the extremities A B, being of the same verticity, mutually abhor and fly one another. For if C be the northern pole of the terrella, A and B are also northern ends; but the ends which are joined to and held at the pole C are both * southern. But if those spikes be a little longer (as, for example, of two digits length) and be joined by force, they adhære together and unite in a friendly style, and are not separated without force. For they are magnetically welded, and there are now no longer two distinct ends, but one end and one body; no less than a wire which is doubled and set up perpendicularly. But here is seen also another subtile point, that if those spikes were shorter, not as much as the * breadth of one digit, or even the length of a barleycorn, they are in no way willing to harmonize or to stand straight up at the same time, because naturally in shorter wires the verticity is stronger in the ends which are distant from the terrella and the magnetick discord more vehement than in long ones. Wherefore they in no way admit of an intimate association and connection.
Likewise if those lighter pieces of iron or iron wires be suspended, hanging, as A and B, from a very fine silk thread, not twisted * but braided, distant from the stone the length of a single barleycorn, then the opposing ends, A and B, being situated within the orbe of virtue above the pole, keep a little away from one another for the same reason; except when they are very near the pole of the stone C, the stone then attracting them more strongly toward one end.
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.
207 ^ Page 132, line 28. Page 133, line 1. nutat.—The editions of 1628 and 1633 both wrongly read mutat.