62 | REPORT—1885. 
mercial requirements, the pentane standard of Mr. Vernon Harcourt— 
since it has no wick and consumes a material of definite chemical com- 
position—when properly defined, is an accurate and convenient standard, 
and gives more accurately than the so-called standard candle an illumi- 
nation equal to that which was intended when the Act was framed. 
Yet the Committee, while desirmg to impress the Board of Trade and 
the public with these views, do not feel inclined at present to recommend 
the adoption of any standard for universal adoption until, further in- 
formation on radiation having been obtained from experiment, they may 
learn whether or not it may be possible to propose an absolute standard, 
founded, like electrical and other standards, on fundamental units of 
measurement——a standard which, for these reasons, would be acceptable 
to all civilised nations. They are, however, inclined to look upon the 
pentane lamp as an accurate means of obtaining an illumination to replace 
the so-called standard candle. 
Proposed Experimental Researches. 
Radiation is measured as a rate of doing work, and consequently 
radiation might be measured in watts. The illumination (or luminous 
effect of radiation) depends partly upon the eye, and is a certain function 
of the total radiation. This function depends upon the wave-length of 
the radiation, or on the different wave-lengths of which the radiation, if 
it be compound, is composed. This function of the radiation perceived 
by the eye is partly subjective, and varies with radiations of different 
wave-lengths and with different eyes. Thus the illumination cannot, like 
the radiation, be expressed directly in absolute measurement. But the 
connection between the illumination end the radiation can be determined 
from a large number of experiments with a large number of eyes, so as to 
get the value of the function for the normal human eye. This function, 
however, is constant only for one source of light, or, it may be, for sources 
of light of the same temperature. It appears, then, that, in the first 
instance at least, a standard should be defined as being made of a definite 
material at a special temperature. 
The energy required to produce a certain radiation in the case of a thin 
filament of carbon or platinum-iridium heated by the passage of an electric 
current can be easily measured by the ordinary electric methods, and the 
radiation may be measured by a thermopile or a bolometer, which itself 
can be standardised by measuring the radiation from a definite surface 
at 100°C., compared with the same at 0°C. The electric method 
measures the absorption of energy; the thermopile measures the total 
radiation. These two are identical if no energy is wasted in convection 
within the glass bulb of the lamp, by reflection and absorption of the glass, 
and by conduction from the terminals of the filament. Captain Abney and 
General Festing have come to the conclusion that there is no sensible loss 
from these causes. The Committee propose to investigate this further. 
This constitutes a first research. 
No research is necessary to prove that with a constant temperature of 
a given filament the luminosity is proportional to the radiation, because 
each of these depends only upon the amount of surface of the radiating 
filament. It will be necessary, however, to examine whether with 
different filaments it be possible to maintain them at such temperatures as 
shall make the illumination of each proportional to the radiation. This 
will be the case if spectrum curves, giving the intensity of radiation in 
