ring; but if it be a little inclined from the perpendicular, the ring will move about its axis to the sides of the magnet on which is the acute angle.

Exp. The current in the parts of the conductor, and those of the adjacent sides of the wire, are in the same direction, hence, the magnet is attracted on all sides, and equally, when its center coincides with that of the ring, and its axis is perpendicular to its plane, but when oblique the attraction is strongest at the two acute angles, and hence the ring will approach the magnet on those sides.

PH. 76. Things being as in the last, if now the poles of the magnet be inverted, while the course of the current remains the same, the ring will place itself so that its plane shall be at right angles to the axis of the magnet, whatever angle it makes with it at first.

Exp. Here there is evidently repulsion at all the four angles, because the currents are opposite, but when the magnet is inclined to the plane of the ring, the repulsion is greatest in the acute angles, hence the reason of the effect is evident, since there can be no permanent equili← brium, but when in the above mentioned position.

PH. 77. If two wires from the opposite ends of a voltaic circuit with large plates be placed perpendicularly in a basin of mercury, and if a powerful magnet be held above or below either of the wires, the mercury revolves, forming a vortex about the wire as an axis; the velocity is increased, if the opposite poles of two magnets be placed, the one above, and the other below the wire.

Exp. While the magnet is held in a line with the wire, or still more when the opposite poles of two magnets are placed the one above, and the other below, in the direction of the perpendicular part of the wire, the spiral or vertical current, which revolves about the magnet or magnets, (ph. 41, sect. ix.) communicates its motion to the fluid mercury, as it does in other cases to the air, and this agrees perfectly with ph. 117, and 68, obs. sect. ix, where similar motions are indicated.

PH. 78. Things being as in the last, if now the magnet be held above the mercury in the middle between the wires, the circular motion ceases, and currents are produced, the one to the right, and the other to the left of the magnet.

Exp. The magnet being now perpendicular to the current in the mercury between the wires, that current is affected by the revolving current of the magnet, which causes the observed disturbance in that liquid.

PH. 79. If two thick copper wires be cemented into two holes, about 3 inches apart, in the bottom of a shallow glass vessel, the wires being coated with sealing-wax, except at the ends which are to be made flat, and polished; the basin being now filled with mercury to of an inch above the ends of the wires, which are to be placed upright, and connected with the extremities of the battery; it will be found that, when the circuit is completed, the mercury rises to or of an inch above each wire in the form of a cone, and, if the action be strong, waves flow in all directions from the cones, the point of rest being near the center between the wires.

Exp. The repulsion between the quicksilver and the ethereal matter, constituting the electric current, and passing from the positive wire to the mercury, and, again from the mercury to the negative wire, elevates the liquid,

especially as it is necessitated to leave the one wire, and to enter the other at the end; hence the conical elevation is produced, as is the motion of the wire in ph. 1, or as the electric fly is moved; and as it is raised continually, it must needs flow down on all sides, and produce the waving motion.

PH. 80. Things being as in the last, if a powerful magnet be placed some inches above one of the cones, its elevation is diminished, and as the magnet is lowered, the altitude of the cone decreases, till the surface of the mercury becomes flat, and a rotation commences slowly

about the wire.

Exp. The action of the magnet tends to produce a vortex in the mercury as in ph. 77; hence, when the magnet is powerful, it depresses the conical elevation, and produces the circulating motion.

Revolving Motions.

PH. 81. The pole of a magnet will move round a conductor, if not prevented by intervening objects, and the motion will be according to the order pointed out in the preceding phenomena (see ph. 18, and following). Let CD, fig. 101, be a non-conducting vessel containing mercury, with which the wire A communicates, and into which the wire B dips about half an inch. Now, if a magnet sn be attached by a thread s to the wire A, so that its north end n may reach a little above the mercury, while the wire B is fixed; when the connection is

made with the voltaic apparatus, by joining A to the positive, and B to the negative end, the magnet will revolve about B in the direction dénoted by the arrow heads. marked on the circumference of the circle, which is drawn about B the fixed wire.

Exp. Let the experimenter have the apparatus before him, and first let the magnet be between him and the wire B; then the current in B is upward, as shewn in the detached figure by the arrow head on ab, and the parts of the spirally revolving current, in sn, on the side next to B, are in the direction cd: hence there is an attraction in the angles dob, and aoc, and a repulsion in the angles boc, and aod, (ph. 7), therefore the tendency is to bring the magnet sn to the position in which its north pole n may be to the right, and consequently, from the circumstances of the connection, n must move to the right as marked by the nearer arrow head on the circle; but this relation of the currents remains on every side of the wire, as may be seen by placing sn on the farther side of B, and using the detached figure where cd' shews the direction of the parts of the current on the side of the magnet next to the wire; from this it is evident, that a continued revolution must take place, the freedom of the motion being sufficient for that purpose. The above reasoning holds good whether sn be a north, or a south magnet.

PH. 82. If the magnet be so attached that its south pole may be uppermost, its revolution about the wire will be in the opposite direction: also if the direction of the electric current in B be reversed, the motion will likewise be reversed.

Exp. The reason of all these effects is evident, when we observe that the relations of the currents in the magnet and wire become exactly opposite in consequence of these changes.

PH. 83. Conversely, if the magnet be fixed, and the wire free to move, it will revolve about the magnetic pole; if the electric current be from A to B, (fig. 101, sn being supposed upright), the wire B will rotate about n, as marked on the circle, and will be reversed by reversing the magnet, or the current, but will continue the same by reversing both.

Exp. This is a necessary consequence, as will appear from the two last, action and re-action being equal and opposite or the same may be explained by considering the relations of the currents as above.

PH. 84. If both the wire, B, and the magnet, sn, fig. 101, be free to move, they will revolve about each other according to the law above stated, each in a conical surface, of which the common axis is the line joining the wires A and E.

Exp. The three last phenomena render this sufficiently clear.

PH. 85. Let ab, fig. 102, be a section of a channel cut in a board, and w the section of a wire suspended freely by a loop, from an upright fixed on the same board, reaching down so far as to dip a little into some pure mercury put into the cavity ab; SMN a section of a horse-shoe magnet placed on the same board, as in the figure, S being the south, and N the north pole: now if the ends of the voltaic apparatus be connected in such a manner, that it would cause the wire to rotate about the north pole, as denoted by the arrow heads on the circle de, (see ph. 83); the wire will in this case be projected out of the mercury towards a, and, as it falls back, will continue to be repeatedly thrown out, if the position of the magnet, or the course of the current were reversed, the wire would in like manner be projected towards b, but the effect would remain as at the first, if both were reversed.