would be under the same difference of electric pressure —namely, the difference of pressure at the two poles. A little lamp, therefore, placed on any wire represented by a meridian would glow with the same brilliancy as that of its neighbour on another meridian; and, furthermore, according to Kirchhoff's laws, the resistance opposed to the electric flow by the lamps would be enormously diminished by this arrangement, which is termed the multiple circuit. We can obtain a conception of a multiple circuit and its effect in diminishing resistance by considering the flow of water through pipes. If we narrow the diameter of a pipe we increase the resistance to the flow of water; if, however, we should undertake to carry water from one pole of the earth to the other pole by means of pipes, we could accomplish this with the least resistance by employing a number of small pipes of small diameter, situated like meridians of the earth, all connecting with the principal mains at the north and south poles. The greater the number of small meridian pipes the less the resistance to the flow through the large mains.

By means of the carbon filament in an exhausted globe and by the arrangement of the multiple circuit the incandescent system of electric lighting became a commercial success.

CHAPTER V.

FLOW OF ELECTRICITY IN THE EARTH.

In the following chapter let us examine the passage of electricity through the earth; for it is well known that it was discovered in the early days of telegraphy that a return wire between Boston and New York, for instance, could be dispensed with, and that the earth could be used instead of the return wire, thus halving the injurious resistance of the circuit; for it was found that the earth did not oppose any appreciable resistance compared with the total length of the telegraphic circuit.

We can find no analogy between the flow of steam, gas, or water and the case of the return circuit through the earth. In the case of steam, gas, and water, and of all fluids forced through pipes from a powerhouse, nothing returns to the power house if we should connect the pipes to the ground; for the steam would be condensed, the air pressure lost, and the water would soak into the ground. In the case of electricity, however, nothing is lost by connecting the wires leading from the power house or battery to the ground. Indeed, in certain cases a great deal is saved, for the energy of the current is not dissipated into heat along a return wire. We have said that a magnet or compass needle instantly points to a wire through which an electrical current

is passing. It is like the finger of a mute person pointing out a secret. It points to the wire if it is moved along the wire from one earth plate to which the wire may be attached to the power house or battery, and from the power house or battery to the earth plate at the other end of the wire circuit. If placed on the earth between these earth plates and sufficiently far from the overhead wire on the ground, for instance, beneath an ordinary telegraph wire strung on poles the compass or magnet is quiescent and performs its normal task of pointing to the poles of the earth. It gives no evidence of an electric current in the ground; the electricity, so to speak, seems to have leaked away like water. Yet instruments show that the current apparently flows out from the power house in one direction to the ground and returns from the ground to the power house.

If we should take a minature earth-a globe of metal, for instance-several feet in diameter, and run an electric current to what may be called the north pole of such a globe and lead it away from the south pole, we shall find that the current apparently spreads out from the north pole and converges, so to speak, to the south pole. If the globe were 20 feet in diameter very little indication of a current would be obtained around the equator of such a globe. Let us now build up a globe made of steam or water pipes all connected to one main pipe at the north pole, and again at the south pole. We can suppose the pipes to represent the divisions of an orange. When steam or water leaves the main pipe and is divided in its flow equally among the pipes, placed similarly to the divisions of the orange; the amount of flow through any one pipe can be made very small, although the flow through the main pipe leading to the globe is very large. If we should connect any two neighbouring

pipes along the equator so that water or steam could flow from one to the other, we should find that there would be no flow, for the pressure at the two ends of such a connecting pipe is the same; there is no difference of pressure to force the steam or water from one pipe to the other. If, however, we should connect one pipe at a point on the equator-in Africa, for instance -with another pipe at a point corresponding on the globe to New York, there would be a flow in the connecting pipe, for there would be a difference of pressure. In the case of electricity, a telephone will determine whether there is a flow from one portion of the earth's surface to another when we lead an electric current into the earth and out of it. Let us use the telephone at first merely as a detector of an electrical flow, just as we used in the above illustration a pipe connecting two pipes in order to determine whether there is any possibility of a flow of water between them. That it can be so used we can easily ascertain, for we hear a click in the telephone whenever we touch the two wires leading to it to the two poles of an ordinary battery such as is used on bell wires or for medical purposes. If we should hold a telephone to the ear and connect one of its leading wires to the rail of an ordinary electric road and the other to the iron posts which run beside the track, we should hear a click at the moment of making the contact with the iron pole if there is a leakage of electricity from the overhead wire, which is supported by the iron pole and its connections, into the ground. In other words, a difference of electrical level would be shown between the iron post where it enters the ground and the rail. If now we should make a globe of a number of copper wires, insulated from each other and forming the meridians of such a globe, and connect all these

great circles of copper wire together to one wire at the north pole and to another wire at the south pole, and lead a current of electricity into the collection of meridian wires at the north pole and out of the collection at the south pole, we should find that very little current would go through any one wire; and if we should connect our telephone wires with two neighbouring wires anywhere along the equator we should hear no click; there is no flow of electricity between points of the same pressure. If, however, we connect one wire of the telephone at one point on the equator and the other wire at a point on the wire globe corresponding to New York, we should hear a click, for there would be a flow between these points.

From such experiments we see that what we call the electric current flows out in all directions from the point where it enters the earth, and appears to converge again to the point where it leaves the ground to enter the wire and to return to the power house or battery. Perhaps the best illustration of the manner in which the electric current spreads out in the earth is afforded by a method of telegraphing without wires, which I described in the Proceedings of the American Academy of Arts and Sciences, and which has lately been repeated by Prof. Rubens in Berlin, and by Mr. Preece of the London telegraphic system. In my paper I remarked: "The theoretical possibility of telegraphing across large bodies

* My original researches were made between the observatory at Cambridge and the city of Boston, which were connected by a timesignal wire. The current upon this wire was broken by a clock at regular intervals. I found that I could hear the clock-beats a mile away from the wire by connecting a telephone to a wire and by grounding the ends of the wire 500 or 600 feet apart and parallel with the time circuit.