induction-coil, and receiver or receivers, may be
substituted. And if a shunt, S, of 500 ohms placed across the circuit
makes no difference to the talking in the telephones because of the
interposition of the separating condenser, C, it will readily be
understood that a telegraphic system properly "graduated," and having
also a resistance of 500 ohms, will not affect the telephones if
interposed in the place of S. This arrangement is shown in Fig. 3,
where the "graduated" telegraph-set from Fig. 1 is intercalated into
the telephonic system of Fig. 2, so that both work simultaneously, but
independently, through a single line. The combined system at each end
of the line will then consist of the telephone-set, T1, the telegraph
instruments (comprising battery, B1, key, M1 and Morse receiver, R1),
the "graduating" electromagnets, E1, and E2, the "graduating"
condenser, C1, and the "separating" condenser, C2. It was found by
actual experiments that the same arrangement was good for lines
varying from 28 to 200 miles in length. A single wire between
Brussels, Ghent, and Ostend is now regularly employed for transmission
by telegraph of the ordinary messages and of the telemeteorographic
signals between the two observatories at those places, and by
telephone of verbal simultaneous correspondence, for one of the Ghent
newspapers. A still more interesting arrangement is possible, and is
indicated in Fig. 4. Here a separating condenser is introduced at the
intermediate station at Ghent between earth and the line, which is
thereby cut into two independent sections for telephonic purposes,
while remaining for telegraphic purposes a single undivided line
between Brussels and Ostend. Brussels can telegraph to Ostend, or
Ostend to Brussels, and at the same time the wire can be used to
telephone between Ghent and Ostend, or between Ghent and Brussels, or
both sections may be simultaneously used.

[Illustration: Fig. 3]

[Illustration: Fig. 4]

It would appear, then, that M. Van Rysselberghe has made an advance of
very extraordinary merit in devising these combinations. We have seen
in recent years how duplex telegraphy superseded single working, only
to be in turn superseded by the quadruplex system. Multiplex
telegraphy of various kinds has been actively pursued, but chiefly on
the other side of the Atlantic rather than in this country, where our
fast-speed automatic system has proved quite adequate hitherto.
Whether we shall see