of power has been proved to be
experimentally possible. In the rarefied atmosphere at a height of
five or ten miles from the earth's surface, electric discharges of
very high voltage are conveyed without any other conducting medium
than that of the air. By sending up balloons, carrying suspended
wires, the positions of despatch and of receipt can be so elevated
that the resistance of the atmosphere can be almost indefinitely
diminished. In this way small motors have been worked by discharges
generated at considerable distances, and absolutely without the
existence of any connection by metallic conductors. Possibilities of
the exportation of power from suitable stations--such as the
neighbourhoods of waterfalls--and its transmission for distances of
hundreds or even thousands of miles have been spoken of in relation to
the industrial prospects of the twentieth century.

Comparing any such hypothetical system with that of sending power
along good metallic conductors, there is at once apparent a very
serious objection in the needless dispersion of energy throughout
space in every direction. If a power generator by wireless
transmission, without any metallic connection, can work one motor at
a distance of, say, 1,000 miles, then it can also operate millions of
similar possible motors situated at the same distance; and by far the
greater part of its electro-motive force must be wasted in upward
dispersion.

The analogy of the wireless transmitter of intelligence may be
misleading if applied to the question of power. The practicability of
wireless telegraphy depends upon the marvellous susceptibility of the
"coherer," which enables it to respond to an impulse almost
infinitesimally small, certainly very much smaller than that
despatched by the generator from the receiving station. From this it
follows, as already stated, that the analogy of apparatus designed
merely for the despatch of intelligence by signalling cannot safely be
applied to the case of the transmission of energy.

Making all due allowances for the prospects of advance in minimising
the resistance of the atmosphere, it must nevertheless be remembered
that any wireless system will be called upon to compete with improved
means of conveying the electric current along metallic circuits.
Electrical science, moreover, is only at the commencement of its work
in economising the cost of power-cables.

The invention by which one wire can be used to convey the return