r at
Charleston in 1863.

In 1848 the first electric arc lamp used for general lighting was
installed in Paris. It was supplied with current by a large voltaic
cell, but the success of the electric arc was obliged to await the
development of a more satisfactory source of electricity. A score of
years was destined to elapse, after the public was amazed by the first
demonstration, before a suitable electric dynamo was invented. With the
advent of the dynamo, the electric arc was rapidly developed and thus
there became available a concentrated light-source of high intensity
and great brilliancy. Gradually the size was increased, until at the
present time mirrors as large as seven feet in diameter and electric
currents as great as several hundred amperes are employed. The beam
intensities of the most powerful search-lights are now as great as
several hundred million candles.

The most notable advance in the design of arc search-lights was achieved
in recent years by Beck, who developed an intensive flame carbon-arc.
His chief object was to send a much greater current through the arc than
had been done previously without increasing the size of the carbons and
the unsteadiness of the arc. In the ordinary arc excessive current
causes the carbons to disintegrate rapidly unless they are of large
diameter. Beck directed a stream of alcohol vapor at the arc and they
were kept from oxidizing. He thus achieved a high current-density and
much greater beam intensities. He also used cored carbons containing
certain metallic salts which added to the luminous intensity, and by
rotation of the positive carbon so that the crater was kept in a
constant position, greater steadiness and uniformity were obtained.
Tests show that, in addition to its higher luminous efficiency, an arc
of this character directs a greater percentage of the light into the
effective angle of the mirror. The small source results in a beam of
small divergence; in other words, the beam differs from a cylinder by
only one or two degrees. If the beam consisted entirely of parallel rays
and if there were no loss of light in the atmosphere by scattering or by
absorption, the beam intensity would be the same throughout its entire
length. However, both divergence and atmospheric losses tend to reduce
the intensity of the beam as the distance from the search-light
increases.

Inasmuch as the intensity of the beam depends upon the actual brightness
of the light-source, the