Gilbert dwells at length on the inductive action of the earth. He hammers heated bars of iron on his anvil and then allows them to cool while lying in the magnetic meridian. He notes that they become magnetized, and does not fail to point out the polarity of each end. He likewise attributes to the influence of the earth the magnetic condition acquired by iron bars that have for a long time lain fixed in the north-and-south position and ingenuously adds: "for great is the effect of long-continued direction of a body towards the poles." To the same cause, he attributes the magnetization of iron crosses attached to steeples, towers, etc., and does not hesitate to say that the foot of the cross always acquires north-seeking polarity. In a similar manner, every vertical piece of iron, like railings, lamp-posts, and fire-irons, becomes a magnet under the inductive action of the earth. In the case of our modern ships, the magnetization of every plate and vertical post, intensified by the hammering during construction, converts the whole vessel into a magnetic magazine, the resulting complex "field" rendering the adjustment of the compasses somewhat difficult and unreliable. The unreliable character of the adjustment arises mainly from the changing magnetism of the ship with change of place in the earth's magnetic field, the effect increasing slightly from the magnetic equator to the poles. With luminous insight into the phenomena of terrestrial magnetism, Gilbert observes that in the neighborhood of the poles, a compass-needle, tending as it does to dip greatly, must in consequence experience only a feeble directive power. To which he adds that "at the poles there is no direction," meaning, no doubt, that a compass-needle would remain in any horizontal position in which it might be placed when in the vicinity of the magnetic pole. This is precisely the experience of all Arctic explorers, who find that their compasses become less and less active as they sail northward, the reason being that the horizontal component of the earth's magnetic force, which alone controls the movements of the compass-needle, decreases as the ship advances and vanishes altogether at the magnetic pole. When once a high latitude is reached, captains do not depend upon their compasses for their bearings, but have recourse to astronomical observations. In his account of magnetic work carried on in the neighborhood of the magnetic pole, Amundsen says: "At Prescott Island the compass, which for some time had been somewhat sluggish, refused entirely to act, and we could as well have used a stick to steer by."