As we may otherwise express it, a given amount of energy may be furnished by the aid of water, either by sending a large quantity of water through the mains at a low pressure, or a small quantity at a high pressure. The work done, or the effect in energy, depends on the product of quantity and pressure,―e.g., 100 pounds of water per minute at a head of 10 feet is equivalent to one pound per minute under a head of 1000 feet, although it is to be noted that a much smaller pipe would be necessary to convey the one pound than the 100. Exactly the same is true of electricity. One hundred ampères at a pressure of 100 volts is the exact equivalent of 10 ampères at a pressure of 1000 volts. The importance of this point is easily seen. When a current of electricity passes along the copper conductor or wire, which takes the place in this case of the hollow water-main, a certain resistance is experienced analogous to the frictional resistance which accompanies the motion of water through a pipe. With a copper main of a given cross-sectional area this resistance increases within certain limits-in direct proportion to the magnitude of the current,—that is, double the current gives double the resistance, &c. Moreover, this resistance, like friction, is accompanied by the generation of heat; the copper conveying the current becomes hot, and a certain point is easily reached at which the heat so generated is sufficient to injure the insulating material round the metal. Whether or not, however, the heat should be unduly great, all the work done against the resistance is absolutely wasted, for useful purposes, exactly as is the similar work done in pumping water. Practically it is found by most low tension companies that it is on these grounds inadvisable to allow copper mains to carry a current of more than six or seven hundred ampères per square inch of cross-section. As the output of a large station at full load is measured in thousands of ampères, this means that the trunk mains must have in all an area of several square inches, which is, of course, a very expensive matter.

On the other hand, the amount of current which a copper rod will carry is not affected by the pressure at which the current is circulating. Hence there is an obvious primâ facie

advantage in the use of a high pressure. It allows the quantity of current for any given amount of electrical energy to be reduced just in proportion as the pressure is increased, and as the former only affects the size of the mains, the use of high tension-other things being equal-allows the weight of copper in the mains to be reduced very materially. Moreover, by using a pressure so high that the loss of head in forcing the current through the mains is negligible, or at least is proportionately very small, the generating station can be placed at a considerable distance, several miles for example, from the district supplied. In itself this is, of course, not desirable-indeed, it is most undesirable, but naturally there may be places where the erection of a lighting station in the centre of a district which it has to supply is for certain sufficient reasons impossible.

I have said that the pressure of electricity in the customers' houses is a low pressure, and must be everywhere the same, whatever may be the pressure in the mains. Practically it is always about 100 volts. Hence clearly a distribution at high pressure would be impossible if there did not exist some ready means of changing the pressure from high to low at the customers' house. This requirement is met by the transformer or converter, which is at once the blessing and the trouble of the high tension systems. To explain to you the nature of this apparatus, I am sorry to say that I have to leave my hydraulic analogy, which appears to furnish no parallel case, and must take the thing just as it stands.

It is necessary before doing this to look for a moment at the other division between the London companies, namely, the use of continuous and alternating currents, pointing out at the outset that so far as house-lighting goes, there is no difference between them, but that as yet motors have practically only been driven by continuous currents. If you recollect my description of the way in which current is produced by a dynamo, you will see that that current really cannot be completely continuous. It consists of a series of very rapidly occurring successive flows of current. For all practical purposes, however, this current may be treated as continuous, for by a piece of apparatus called a commutator, which

forms part of the dynamo, all the currents are made to flow in the same direction. A really continuous current is only furnished by a battery, but so far as all its leading characteristics and its measurement go, the current from a continuous current dynamo is identical with that furnished by a battery. Such a current, as we shall see directly, cannot be transformed in pressure in the way I have alluded to. It is quite possible, however, to arrange a dynamo-which then requires. no commutator-so that it delivers its current, not continuously, but in what may be called wavelets, swinging first forward, then back. An electric current of this kind is called an alternating current, and is used by all the high tension. companies which I have mentioned in London. Each wave of current is complete in itself, and occupies generally from one-eightieth to one-hundredth of a second, forming a complete cycle of what-in default of a better expressionwe may call a go and return current, and between each pair of waves the current is zero, that is, the flow ceases absolutely, although for an infinitesimally short time.

The "transformer" used with alternating currents is based on the following phenomena:-If two conductors of a suitable kind, and forming part of closed circuits, are placed near together and parallel, but without any contact (that is completely insulated from each other), and a current be passed through the one, at the instant of starting it a current will be found to traverse the other. This latter current (which is called an induced current) does not continue, whether or not the originating current goes on steadily; it is instantaneous only, occurring just at the instant of starting. But at the instant when the first or primary current ceases, another induced current appears in the second wire, this time in the opposite direction to that in which it formerly appeared. If, therefore, the primary current starts and stops eighty times per second, the induced or secondary current will be a similar discontinuous current having the same frequency and therefore available for the same purposes. If the primary and secondary wires be alike, the tension of the two currents as well as their quantity will be the same. By using, instead of a straight wire, a helix of a certain number of turns to form the primary, and

for the secondary another helix containing a different number of turns, the induced current may be made either higher or lower in pressure than the primary. The quantity of current, of course, varies inversely as the pressure, so that the amount of energy transmitted is not altered, except so far as internal losses affect the efficiency of the apparatus. In principle the transformer is simply a pair of coils such as I have mentioned. Remember that there is no electrical contact whatever between the coils, although they may be in very close proximity. The high tension main goes from the station to the primary coil of the transformer and back again to the station without actually entering the house. The low tension main goes from the secondary coil of the transformer round the house circuits. Apart from accidents, therefore, the high tension current can never actually get into the house mains. Figure 1* shows the general arrangement of the circuit and transformers used by the Metropolitan Electric Supply Corporation. A is the central station from which a series of circuits are led through the district. D represents a dynamo, with its two poles connected to one of these circuits. It will be seen that soon after leaving the station both positive and negative wires are looped into rings; the ring may, of course, embrace many streets. B, B, are houses in the district to be supplied with current. There is a transformer, T, and a meter, M, in every house. Wires from the mains enter the transformer, pass through the primary coil, and come out again. The secondary coil is shown separately connected with the lamp circuits of the house. The meter by which the electric energy is measured is on the house circuit. The pressure in the mains is 1000 volts, the pressure in the house circuits may be either 100 volts or 50 volts according to the wish of the consumer. Mr Frank Bailey, the engineer to the company, is of opinion that a 50-volt house circuit is better than a 100volt house circuit, from the view at any rate of the life of the lamps. The system of looped mains shown in the sketch is due to Mr Bailey, and has been adopted so that every house on a circuit can be reached round the loop in two ways, thus

*The same lettering is used in all the figures.

greatly reducing the risk of breakdown through an accident on the mains. One or more circuits may be connected to and worked from the same dynamo at the station, but at present it has not been found possible here to put more than one dynamo on to the same circuit, or, in technical language, to run the dynamos in parallel.

The system adopted by the London Electric Supply Corporation, which has been from the commencement designed and worked out by Mr Ferranti, differs from that just described in very many respects. Its general scheme is sketched in figure 2. This company has, rightly or wrongly, assumed that it was advisable to make its station away from London. Its generating station is on the Thames at Deptford, about six miles from Charing Cross. Current is there generated by large dynamos at a pressure of 2500 volts. It is transformed at once up to 10,000 volts, and transmitted to London at that pressure by specially insulated mains of most ingenious construction. These mains lead to two or three sub-stations in London itself, one in Bond Street, one near Charing Cross, &c., which contain transformers only, and where the current is transformed down to 2500 volts again, and distributed at this pressure through the streets. In the customers' houses it is transformed down to 100 volts, just as in the last case. It is intended ultimately to generate current at the full pressure of 10,000 volts so as to save the first transformation. There can be no doubt that with such appliances and knowledge as we possess at present the drawbacks of having three transformers are very great indeed. Whether or not they are more than counterbalanced by the advantage of having a station down the river is a matter which must finally be decided by the result of practical working.

The system used by the House-to-House Company does not differ in essentials from that of the Metropolitan Electric Corporation, except that the pressure in the mains is 2500 volts instead of 1000.

I have said above that the transformer was the trouble as well as the blessing of the alternating current system. Whilst it enables a saving to be made apparently throughout the

VOL. XIII.

H