Action and re-action are equal and opposite, hence if the needle is fixed and the wire free the current will move round the magnet; and if both are free they will circle round each other. Applying the above rule we shall find that when the north pole moves from left to right the current moves from right to left. Ampere of Paris, following Oersted, promptly showed that two parallel wires carrying currents attracted each other when the currents flowed in the same direction, and repelled each other when they flowed in opposite directions. Thus, in figure 32, if A and B are the two parallel wires, and A is mounted on pivots and free to move in liquid "contacts" of mercury, it will be attracted or repelled by B according as the two currents flow in the same or in opposite directions. If the wires cross each other at right angles there is no attraction or repulsion. If they cross at an acute angle, they will tend to become parallel like two compass needles, when the currents are in one direction, and to open to a right angle and close up the other way when the currents are in opposite directions, always tending to arrange themselves parallel and flowing in the same direction. These effects arise from the circular lines of force around the wire. When the currents are similar the lines act as unlike magnetic poles and attract, but when the currents are dissimilar the lines act as like magnetic poles and repel each other. Another important discovery of Ampere is that a circular current behaves like a magnet; and it has been suggested by him that the atoms are magnets because each has a circular current flowing round it. A series of circular currents, such as the spiral S in figure 33 gives, when connected to a battery C Z, is in fact a skeleton ELECTRO-MAGNET having its north and south poles at the extremities. If a rod or core of soft iron I be suspended by fibres from a support, it will be sucked towards the middle of the coil as into a vortex, by the circular magnetic lines of every spire or turn of the coil. Such a combination is sometimes called a solenoid, and is useful in practice. When the core gains the interior of the coil it becomes a veritable electromagnet, as found by Arago, having a north pole at one end and a south pole at the other. Figure 34 illustrates a common poker magnetised in the same way, and supporting nails at both ends. The poker has become the core of the electromagnet. On reversing the direction of the current through the spiral we reverse the poles of the core, for the poker being of soft or wrought iron, does not retain its magnetism like steel. If we stop the current altogether it ceases to be a magnet, and the nails will drop away from it.