n, the diaphragm will be merely
flexed and return to its original form without producing a sound. If,
therefore, there be placed in the same circuit a regular telegraph relay
and a special telephone, an operator may, by manipulating a key, operate
the relay (and its sounder) without producing a sound in the telephone,
as the makes and breaks of the key are far below the limit of
audibility. But if through the same circuit, by means of another key
suitably connected there is sent the rapid changes in current from an
induction-coil, it will cause a series of loud clicks in the telephone,
corresponding to the signals transmitted; but this current is too weak
to affect the telegraph relay. It will be seen, therefore, that this
method of duplexing is practiced, not by varying the strength or
polarity, but by sending TWO KINDS OF CURRENT over the wire. Thus, two
sets of Morse signals can be transmitted by two operators over one
line at the same time without interfering with each other, and not only
between terminal offices, but also between a terminal office and any
intermediate office, or between two intermediate offices alone.

III

AUTOMATIC TELEGRAPHY

FROM the year 1848, when a Scotchman, Alexander Bain, first devised a
scheme for rapid telegraphy by automatic methods, down to the beginning
of the seventies, many other inventors had also applied themselves to
the solution of this difficult problem, with only indifferent success.
"Cheap telegraphy" being the slogan of the time, Edison became arduously
interested in the subject, and at the end of three years of hard work
produced an entirely successful system, a public test of which was made
on December 11, 1873 when about twelve thousand (12,000) words
were transmitted over a single wire from Washington to New York. in
twenty-two and one-half minutes. Edison's system was commercially
exploited for several years by the Automatic Telegraph Company, as
related in the preceding narrative.

As a premise to an explanation of the principles involved it should be
noted that the transmission of telegraph messages by hand at a rate of
fifty words per minute is considered a good average speed; hence, the
availability of a telegraph line, as thus operated, is limited to this
capacity except as it may be multiplied by two with the use of
the duplex, or by four, with the quadruplex. Increased rapidity of
transmission may, however, be accomplished by automatic methods, by
mea