532 
CAPTAIN W. DE W. ABNEY AND MAJOR-GENERAL E. R. RESTING 
with each other, and the angular dimensions of the patch had not to be taken 
into account. 
An extension of the measurements to embrace a part of the physiological aspect of 
colour has, however, necessitated a slight modification of the apparatus wfith which a 
new series of observations have been made. The length of the spectrum, before 
re-combination, has been more than doubled by using a lens of greater focal length than 
formerly for collecting the rays proceeding from the prisms. The size of the white 
surface on which the shadows are received has been reduced to inch square, and the 
observations now recorded were made from a distance of 4 feet from it. This allows the 
image of the whole of the patch, when viewed direct, to fall on the middle of the yellow 
spot of the eye, which occupies a central position in the retina, and has an approximate 
angular aperture of from 6° to 8°. As the absorption of the yellow spot diminishes 
towards its boundary, it follows that, within certain limits, the smaller the patch that 
is viewed the greater will be the loss of luminosity in that part of the spectrum where 
the absorption takes place. From observations made in the manner shortly to be 
described, it was found that, for our eyes, there was no sensible difference in the 
results when the two shadows fell on the white square of l|~inch side, or a square of 
1-inch side, but that, if the side were increased to 2 inches, the measures differed 
slightly but unmistakably. This will account for the fact that the original luminosity 
curve slightly differs from that now recorded for the centre of the eye, as part of the 
image of the larger patch must have fallen on the less absorbing part of the yellow 
spot. It is not quite apparent why the eye should not distinguish between the 
differing luminosities of the different parts of the shadows, but it is probable that the 
average luminosity was observed. 
Table I., Col. IV., gives the observed measurements, and fig. 33 gives the curve A 
plotted from them. 
Equalisation of the luminosities of the coloured and white shadows was effected by 
opening and closing the rotating sectors (which were described in the former paper) 
in the white beam of light, and confirmation of the measurement was obtained by 
setting the sectors at fixed angles, thus cutting off definite proportions of white light 
and shifting the slide carrying the slit which traverses the spectrum till equality of 
illumination was obtained at each ancle so set. 
O 
From our experience we believe that the most accurate measurements are those 
% 
made by altering the angular aperture of the sectors during rotation ; as to produce a 
certain difference of luminosity a greater motion of the hand is required on the lever 
of the rotating sectors than on the slit in the spectrum. Near the place of maximum 
luminosity the latter plan fails, as pointed out in our previous paper. 
§ XLV .—Absorption of the Yello w Spot. 
Though it was not the first inquiry which was undertaken, it will be well thus 
