-ARTIFICIAL ILLUMINATION, 171 
I direct your special attention to the annual expenditure necessary 
in order to maintain thirty 16 candle electric lamps. You see that it 
amounts to £65. Good ordinary gas burners would give us an equal 
light annually for £40 and “ Welsbach” incandescent lamps for £14. 
If, however, the electricity had been supplied at 6d. per unit the annual 
cost would have been £78. 
With a larger installation of 300 lamps, the annual cost ot the 
electric light made by our own gas engines would be £367, whereas 
ordinary gas burners would cost us somewhat more, or £400, and the 
“ Welsbach ” incandescent would cost only £140. Generally we may 
say that, with large installations, electricity, locally generated, can beat 
the gas if ordinary burners are used, but it seems still to fail as a 
competitor in the point of economy with the modern “ Welsbach ” 
system. 
aad now allow me to conclude with a few words on the competition 
of electricity and gas and the prospects of its ultimate issue. The 
special advantage hitherto claimed, and justly claimed, for electric 
lighting has always been in its non-pollution of the atmosphere. But 
now that we can get a brilliant gas-light, the deleterious products of 
which are so small as to be almost negligable, the special pre-eminence 
of electricity in this respect exists only in a much smaller degree. 
Electric incandescent lighting, except in such special cases as where, 
for instance, cheap water power is available, is, as we have seen, much 
- more expensive than gas, and there seems little prospect of its being 
cheapened to any great extent. The best combination of engine and 
dynamo yet constructed has not succeeded in transforming more than 
one-tenth of the heat value of the coal used into current energy, and only 
five per cent. of this is transformed into light in the lamps. It would 
appear at first sight that there is here a large and promising margin 
for improvement. As, however, the efficiency of a modern dynamo is 
already as high as 95 per cent. little advance can be looked for here, 
and the question of the cheapening of the electric light resolves itself, 
therefore, mainly into two problems. Firstly, one for the engineer, to 
get more horse-power out of a given quantity of fuel, and secondly, 
one for the physicist to get a better refractory filament which will 
transform a larger proportion than five per cent. of the current energy 
into light. Practical considerations, and a study of thermo-dynamic 
laws alike seem to set a very narrow limit to the advance we may ex- 
pect in the production of horse-power from fuel, and we hardly dare 
hope for a much higher duty than that we now get from the very best 
engines, viz, 1 horse-power for every 14lbs. of coal consumed per hour. 
More hope lies, I think, in the direction of improvements in the lamps 
themselves, and although little or no advance has been made in the 
last five or six years in this respect, yet a much better light-radiating 
filament than the present carbon thread may some day be discovered. 
Side by side, however, with any possible progress in this latter 
- direction, we may expect improvements in the toughness and light- 
radiating power of the mantles used in incandescent gas lighting. 
Marked advance in this respect has been made of late, and I cannot 
help thinking that the incandescent gas system will easily retain the 
