240 ELECTRIC LIGHTING. 



lowered from 36 to 25 of the hydrometer will no longer act 

 with sufficient energy to yield the voltaic arc under advan- 

 tageous conditions. It is also necessary to take into account 

 the loss of mercury, the somewhat greater consumption of 

 zinc than theory indicates, and the consumption of the car- 

 bons between which the arc passes, the value of which is 

 2 francs 50 centimes per lineal metre. "The cost per hour 

 then amounts/' says Becquerel, " to 3 francs for the 60 cells, 

 or about 5 centimes per hour for each cell. However, the 

 cost for a given luminous intensity is not the same at the be- 

 ginning and at the end of an experiment ; this results from 

 the diminution in the electric intensity of the cells, that is to 

 say, from the change in the composition of the liquids they 

 contain. It is nevertheless one result which ought to be 

 stated, as it facilitates the inquiry into the price of working 

 a Bunsen battery ; it is, that for every i franc's-worth of zinc 

 the other materials may be estimated at ifr. 50, so that the 

 total cost cannot be less than 2fr. 50." 



A curious remark made by Becquerel is that the luminous 

 intensity diminishes much more rapidly than the intensity of 

 the current, and this depends upon the fact that the luminous 

 intensity, being a function of the quantity of heat disengaged, 

 must, like it, vary as the square of the quantity of electricity 

 which passes through the circuit in a given time, according 

 to Joule's law. (See page 6.) 



" We see," Becquerel concludes, " that according to the deter- 

 minations I have given, having regard merely to the cost of the 

 substances consumed, and without including labour, the electric 

 light would, light for light, be four times dearer than gas-light- 

 ing at the selling price of gas in Paris. It would be the same 

 as that of lighting by oil, and a quarter of that of lighting by 

 candles. But if we estimate the labour required to superintend 

 the apparatus, to get it ready, and to renew the batteries, &c., the 

 cost would increase by at least half as much again the amount 

 mentioned above." 



According to experiments made at Lyons during 100 hours 



