In the first place, the lighting may be effected in one of three ways—(1) by primary batteries; (2) by dynamo-machines; or (3) by a combination of dynamo-machines and secondary batteries. The expense of working with primary batteries is altogether prohibitory, except in the case of very small installations; while secondary batteries have not yet been made a practical success; so that the second method mentioned above is the only one at present in the field. In the second place, a distinction must be made between isolated installations and a general system of lighting from central stations. Up to the present time, nearly all the lighting by electricity has been effected by isolated installations. If every man requiring one hundred or even several hundred lights were to set up his own gas-works and supply himself from them, the cost of lighting by gas would be enormously increased. Hence it is manifestly unfair to compare the cost of electric light obtained from isolated installations with gas obtained from gas-works supplying many thousands of lights; yet this is being constantly done. Central stations supplying at least, say, ten thousand lights, and gas-works on an equal scale, must be compared in order to arrive at a true estimate of the relative cost of electricity and gas. Several such extended installations are now being erected in London and elsewhere. With improved generating apparatus, and above all, with improved lamps, it is confidently anticipated that the electric light will eventually be cheaper than gas. Even if dearer than gas, it will be largely used for lighting dwelling-houses, theatres, concert-halls, museums, libraries, churches, shops, show-rooms, factories, and ships; while perhaps gas may long hold its own as the poor man’s friend, since it affords him warmth as well as light.