514 ELECTRO-MAGNETISM the currents is the same in the whole series. If we further suppose that the column is stand- ing on end, and that this motion is contrary to that of the sun and contrary to that of the hands of a watch when placed face upward, such ar- rangement will represent the hypothetical mag- in-t <>t' Ampere, in which the north end, or that which turns to the north, is uppermost, and consequently the south pole undermost. If these postulate's be granted, instead of loading the memory with an almost infinite variety of disconnected facts, we shall have at once a gen- eralization from which all the phenomena can be deduced at pleasure in a series of logical corollaries. If this theory be true, or if it be even an approximation to the truth, it will fol- low that if currents of electricity be transmitted through an arrangement of the kind we have described, the phenomena of ordinary magnet- ism will be exhibited ; and this anticipation will be realized if we coil a piece of copper wire covered with silk into the form of a corkscrew spiral, forming a cylinder 8 or 10 in. long, and if the two projecting ends not included in the spiral be passed backward through the cyl- inder and made to project from the middle at right angles to the length of the cylinder on op- posite sides. If this cylinder, the several spires of which will represent the pieces of money, be supported horizontally, so as to turn freely as a magnetic needle moves on its pivot, it will take a north and south position when a power- ful current of galvanism is transmitted through the wire. Nay, more, another cylinder formed of like spires through which a current of gal- vanism is passing will act upon the first precisely as a magnet would act upon another magnet. Indeed, so long as the galvanic current is pass- ing through this helix or spiral, it exhibits all the properties of an ordinary magnet ; but they immediately disappear when the current is in- terrupted. To deduce from his theory the al- most infinite number of facts which it involves, Ampere first considered the action of currents on currents. Starting with the hypothesis that the attraction and repulsion were inversely as the square of the distance between the elemen- tary parts or smallest portion of the currents, he deduced mathematically the consequence that the force of a current of considerable length act- ing on a single element of a current would vary inversely as the simple distance ; and this he was enabled to verify by experiment by suspending a bent wire through which & current was pass- ing so as to be free to oscillate under the influ- ence of a single element, which was ingeniously effected by doubling a piece of covered wire in the middle of its length, thus > . When a cur- rent was passed through this double wire, the portion of it which went to the point of bend- ing and that which came from it neutralized each other, and the residuary effect therefore was that of a single point, which gave a result exactly in conformity to the deduction from the theory. After proving experimentally this fundamental principle, he was enabled by mathematical reasoning, principally of a sim- ple character, to deduce the resultant action of the most complex forms of conductors upon conductors. Among many others, the follow- ing important deductions immediately flow from the premises assumed. If a current of electricity be sent in the direction from A to c B through a straight conductor, A B, of in- definite length, placed B for example horizontal- ly, and a current be sent downward through a terminated conductor, CD, perpendicular to the former, the latter conductor will be impelled parallel to itself along the length of the horizontal conductor. This effect will be due to the fact that on the right side of the short conductor the elements of the two currents are moving in opposite direc- tions; the current in the short wire is ap- proaching the point F, while the current in the horizontal wire is moving from this same point, and hence on this side repulsion will take place ; while on the left-hand side of the short wire the two currents are moving toward the same point, and therefore attraction will be exhibited ; and under the influence of these two forces, the short conductor will move par- allel to itself from right to left along the hori- zontal conductor. If the direction of the cur- rent in either of the two conductors be re- versed, the motion of the short conductor will also be reversed. If, instead of the short con- ductor, one in the form of a ring be freely sus- pended over, the long conductor, with the plane of the latter across the former, the cur- rent passed through this will ascend on one side of the ring and descend on the other. Therefore, the one side will tend to move to the right and the other to the left, and the re- sultant action will be to bring the plane of the ring parallel to the horizontal current; in which case the current in the lower part of the ring will be moving in the same direction as the current in the long wire. Now, since, ac- cording to the theory of Ampere, magnetism depends upon currents of electricity, it follows that the magnetism of the earth results from currents of electricity moving continually from east to west. Hence, if a conductor be bent into the form of a ring or hoop, and freely sus- pended, it will arrange itself east and west. To insure the success of this experiment, the hoop should be formed of a long wire covered with silk and coiled into the form of a ring so as to multiply the actions. Such a ring may be considered as one of the disks represented by the shillings in the hypothetical magnet; and since each disk making up the whole length of the rod would be similarly acted upon by the currents of the earth, the axis of the rod would assume a north and south direction if left free to move, thus affording an explanation of the fact, so long considered an ultimate one, of the directive property of the needle. Let us return again to the action of the long horizon-
