90 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 13, No. 5 
lated can be checked by more direct measurements. It is, therefore, 
highly desirable that when a standard visibility curve is adopted it 
be made to fulfill this condition. In order to do so it may possibly 
be necessary to adjust the experimentally determined visibility curve 
to a slight extent. Whether this will be necessary or not can only 
be determined by comparing results calculated from various visibility 
curves with those established by direct measurement. As has been 
repeatedly indicated above, results obtained by all methods depend 
to some extent on conditions which have to be chosen more or less 
arbitrarily. It is to be hoped that agreement can be reached on the 
specification of conditions such that consistent results can be obtained 
whether one uses the equality-of-brightness photometer, the flicker 
photometer, or spectral measurements in combination with a visibility 
curve. There appears to be no doubt that conditions can be so 
specified as to bring about this condition so far as comparison of 
incandescent radiators at different temperatures is concerned. For 
instance, if the standard visibility curve be adjusted to make it agree 
with the usual equality-of-brightness photometer as now used for 
these measurements, flicker values can readily be brought into agree- 
ment by correcting them to a “normal” slightly different from the 
results which that instrument would give when used by the average 
observer. There would appear to be some justification for adopting 
as one of the basic conditions the use of the small central portion of 
the retina which is actually used in observation of details. If this 
were done it is probable that all three of the photometric methods 
mentioned could be brought into very close agreement. 
If such a solution for the fundamental problem of comparison of 
brightness can be obtained, and the black body is made to furnish 
a standard reference source, our system of assigning values to lights 
of every kind will probably be in as satisfactory a condition as the 
intrinsic difficulties of the subject will ever permit. 
