they said they were, on the trackless ocean by their only trusty guide. Columbus calmed their fears by saying that the needle did not turn to the polar star, but to some fixed and invariable point near it. This explanation, born of inspiration, quieted the sailors, who marveled much at the Admiral's great astronomical knowledge. Gilbert rightly states that the declination changes with place, but he slips into error when he says: "As the needle hath ever inclined towards east or towards 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."
We know, however, that for any given place this angle is continuously, though slowly, changing. Some of these changes require centuries for the completion of their cycle, and are therefore called 'secular'; others require but a year, and are termed 'annual'; whilst others run their course in the space of a day and are known as 'diurnal.' Though these periodic changes in the declination have been established by careful and prolonged observations, we can not say that they are yet satisfactorily accounted for.
The dip of the needle was also familiar to Gilbert, having been first observed in 1576 by Robert Korman. Our philosopher illustrates this phenomenon by balancing a piece of steel so that it remains exactly
Fig. 3. Gilbert's Terella showing the Behavior OF A Dipping-needle at its Poles, Equator and other Intervening Places.horizontal when unmagnetized, and by observing that the moment it is stroked by a magnet its north seeking end dips 'as low as the fulcrum on which it is supported permits.' Gilbert moves a needle over his terella, and finds (1) that the dip is 0° on the equator, (2) that it gradually increases with the latitude being 90° at either pole.
He extends this experiment from the terrella to the earth itself, and even devises an instrument for determining the latitude of any place on land or on sea in the thickest weather and in the darkest night, 'without the help of sunne, moone, or starre.' In this, however, he was wrong. For he assumed the isoclinic lines to be circles running parallel to the magnetic equator, which he erroneously supposed to coincide with the geogrpahical equator.
Gilbert recognized that the earth exerts on a freely movable needle a force that gives it direction and not a motion of translation. He illustrates this by floating a needle on a cork and observing that it points
