effects so well calculated to deceive, if not fairly conditioned. As
we have seen, it is claimed on excellent authority that the equivalent
of 4,000 candles appeared in an arc by expending 40,000 foot pounds of
energy at the generator, but with everyday conditions it is at present
idle to expect such efficiency. Commercially we can give by our own
system 3,000 candles for 40,000 foot pounds absorbed; this may be done
for an indefinite length of time and leave nothing to be desired on
the score of steadiness. Unfortunately there is no unit of photometric
measurement generally recognized in this country, each electrician
having so far adopted one to suit his own convenience; but in making
the foregoing statement I wish it to be understood that our efficiency
would appear still greater if measured by some of the methods now
employed. For our own satisfaction we have endeavored to be at least
approximately accurate, at the same time wishing to avoid the
affectation of extreme precision, such, for example, as adding twenty
or thirty candles to measurements of so many thousands, and we are
satisfied that the most critical expert tests will prove our claim to
be within the mark. The limit of subdivision is only reached when the
difficulty of further increasing the electromotive force of the
machines, involving great care in insulation and a host of other
troubles arising, so to speak, at very high pressure, is balanced by
the objections to working in multiple arc; this appears to occur now
at something below 40 lights, but will in all probability be greatly
extended within a short time. The machines are so constructed that the
local currents, usually productive of dangerous heating, are turned to
useful account, so that the point where radiation exceeds production
is soon reached, and provided the machines are not speeded beyond the
proper limit, they may be run continuously without the slightest
indication of lost vitality. I need scarcely remind you that this is a
most important feature, and by no means a common one.