126 
SIR W. DE W. ABNEY AND PROF. W. WATSON ON 
fig. 27. It will be observed that the two observers, one of whom belongs to class I. and 
the other to class II., get entirely different results. In the case of B., the ratio of the 
foveal to the parafoveal intensities for equal brightness does not differ greatly from 
unity and is very nearly constant throughout the spectrum, showing that the sensitive¬ 
ness of the retina at the fovea and at 2'5 degrees from the fovea is nearly the same, 
a result which is confirmed by the threshold measurements. In the case of W. there 
is a very marked increase in the sensitiveness at 2‘5 degrees as we go towards the 
violet, so that at SSN 5'9 (400 /xm) the central illumination has to be 27 times the 
S SJV 
Fig. 27. 
brightness of the peripheral for the brightness to appear the same. Observer A. 
obtains similar results to those obtained by B., except that his ratio is slightly higher 
at SSN 30 ; the maximum value of the ratio is, however, only 2. B. also made 
observations using a lens to view the illuminated surfaces which caused the image of 
the edge of the annulus to fall at 6'5 degrees from the fovea. The ratio was again 
practically constant, the maximum value being 1'29 at SSN 6. 
In the above series only a single intensity was used for each colour, and W. then 
made sets of observations with different intensities in order to study the manner 
in which the relative sensitiveness varies with intensity. The results obtained are 
