electric light at Menlo Park around the
Christmas period of 1879. He realized, however, that the enormous
investment for copper would militate against the commercial adoption of
electric lighting on an extended scale. His next inventive step
covered the division of a large city district into a number of small
sub-stations supplying current through an interconnected network of
conductors, thus reducing expenditure for copper to some extent, because
each distribution unit was small and limited the drop.

His next development was the radical advancement of the state of the art
to the feeder system, covered by the patent now under discussion.
This invention swept away the tree and other systems, and at one bound
brought into being the possibility of effectively distributing large
currents over extended areas with a commercially reasonable investment
for copper.

The fundamental principles of this invention were, first, to sever
entirely any direct connection of the main conductors with the source of
energy; and, second, to feed current at a constant potential to central
points in such main conductors by means of other conductors, called
"feeders," which were to be connected directly with the source of energy
at the central station. This idea will be made more clear by reference
to the following simple diagram, in which the same letters are used as
before, with additions:

In further elucidation of the diagram, it may be considered that the
mains are laid in the street along a city block, more or less distant
from the station, while the feeders are connected at one end with the
source of energy at the station, their other extremities being connected
to the mains at central points of distribution. Of course, this system
was intended to be applied in every part of a district to be supplied
with current, separate sets of feeders running out from the station to
the various centres. The distribution mains were to be of sufficiently
large size that between their most extreme points the loss would not
be more than 3 volts. Such a slight difference would not make an
appreciable variation in the candle-power of the lamps.

By the application of these principles, the inevitable but useless loss,
or "drop," required by economy might be incurred, but was LOCALIZED IN
THE FEEDERS, where it would not affect the uniformity of illumination
of the lamps in any of the circuits, whether near to or remote from the
station, because any vari