ners of large radiating
surface for their lamps, but he realized the utter futility of such
devices. The tremendous problems of heat and the prohibitive quantities
of copper that would be required for conductors for such lamps would be
absolutely out of the question in commercial practice.

He was convinced from the first that the true solution of the
problem lay in a lamp which should have as its illuminating body
a strip of material which would offer such a resistance to
the flow of electric current that it could be raised to a high
temperature--incandescence--and be of such small cross-section that it
would radiate but little heat. At the same time such a lamp must require
a relatively small amount of current, in order that comparatively small
conductors could be used, and its burner must be capable of withstanding
the necessarily high temperatures without disintegration.

It is interesting to note that these conceptions were in Edison's mind
at an early period of his investigations, when the best expert opinion
was that the subdivision of the electric current was an ignis fatuus.
Hence we quote the following notes he made, November 15, 1878, in one of
the laboratory note-books:

"A given straight wire having 1 ohm resistance and certain length is
brought to a given degree of temperature by given battery. If the same
wire be coiled in such a manner that but one-quarter of its surface
radiates, its temperature will be increased four times with the
same battery, or, one-quarter of this battery will bring it to the
temperature of straight wire. Or the same given battery will bring
a wire whose total resistance is 4 ohms to the same temperature as
straight wire.

"This was actually determined by trial.

"The amount of heat lost by a body is in proportion to the radiating
surface of that body. If one square inch of platina be heated to 100
degrees it will fall to, say, zero in one second, whereas, if it was at
200 degrees it would require two seconds.

"Hence, in the case of incandescent conductors, if the radiating surface
be twelve inches and the temperature on each inch be 100, or 1200 for
all, if it is so coiled or arranged that there is but one-quarter, or
three inches, of radiating surface, then the temperature on each inch
will be 400. If reduced to three-quarters of an inch it will have on
that three-quarters of an inch 1600 degrees Fahr., notwithstanding
the original total amount was but 1200, because th