the office of an eminent electrician. The light was that known as
incandescent--a filament of carbon raised to a light-emitting heat in
vacuo. The exclusion of the air is necessary to prevent the otherwise
rapid destruction of the carbon by combination with oxygen. At the
time of my visit there were 62 lamps in circuit. According to their
statement each lamp was of 16-candle power--I accept their statement
as correct; this will give us an aggregate of 992 candles. The
generator was vitalized by an engine rated by the attendants in charge
at 6-horse power. I found that it was a 5×7 cylinder, working with
very little expansion 430 revolutions per minute, with 90 pounds of
live steam, in a boiler not 15 feet from the engine. I have every
reason to believe that the steam was delivered at the cylinder with an
almost inappreciable loss on 90 pounds. Under those conditions I think
it is perfectly fair to assume (you have the data, so that you can
calculate it afterwards) that 750,000 foot pounds were consumed in
producing those 60 lights, aggregating 992 candles. In the kind of
engine they had, 750,000 foot pounds requires a consumption of about
100 pounds of coal per hour. It was an ordinary high speed engine.
That 750,000 foot pounds, I assume, required 100 pounds of coal. That
is the only weak point in my data; I do not know that to be true; but
I never saw an engine of that form yet capable of delivering 1-horse
power with less consumption than four to five pounds of coal per horse
power per hour. I want to be as fair as I can in the matter. I wish to
compare this, as they have taken particular pains to compare it, with
gas, at the present cost of gas.

The hundred pounds of coal will produce 400 feet of gas; 400 feet of
gas will evolve the effect of 1,500 candles. So you see the position
we are in. In consuming that coal directly by destructive distillation
you can produce 1,500 candles light; by converting it into power, and