ally useful.

Probably the earliest medium chosen to carry wireless speech was light
rays. A microphone transmitter was arranged so that the vibrations
of the voice would affect the stream of gas flowing in a sensitive
burner. The flame was thus thrown into vibrations corresponding to the
vibrations of sound. The rays from this flame were then directed by
mirrors to a distant receiving station and there concentrated on
a photo-electric selenium cell, which has the strange property of
varying its resistance according to the illumination. Thus a telephone
receiver arranged in series with it was made to reproduce the sounds.

This strange, wireless telephone was so arranged that a search-light
beam could be used for the light path, and distances up to three miles
were covered. Even with this limited range the search-light telephone
had certain advantages. Its message could be received only by those in
the direct line of the light. Neither did it require aerial masts
or wires and a trained telegrapher who could send and receive the
telegraph code. It was put to some use between battle-ships and
smaller craft lying within a radius of a few miles. The sensitive
selenium cell proved unreliable, however, and this means of
communication was destined to failure.

The experimenters realized that future success lay in making the ether
carry telephonic currents as it carried telegraphic currents. They
succeeded in establishing communication without wires, using the same
antenna as in wireless telegraphy, and the principles determined are
those used in the wireless telephone of to-day. The sending apparatus
was so arranged that continuous oscillations are set up in the ether,
either by a high-frequency machine or from an electric arc. Where
set up by spark discharges the spark frequency must be above twenty
thousand per second. This unbroken wave train does not affect the
telephone and is not audible in a telephone receiver inserted in the
radio receiving circuit. But when a microphone transmitter is inserted
in the sending circuit, instead of the make-and-break key used for
telegraphy, the waves of the voice, thrown against the transmitter
in speaking, break up the waves so that the telephone receiver in the
receiving circuit will reproduce sound. Here was and is the wireless
telephone. Marconi and many other scientists were able to operate
it successfully over comparatively short distances, and were busily
engaged in extending i