116 
SIR W. DE W. ABNEY AND PROF. W. WATSON ON 
In fig. 18 are given the foveal extinctions for the observers of class II. multiplied 
by the rod luminosities. Except in the orange and red, where we have chiefly to do 
with cone vision, the curve for observer B. is a horizontal straight line, showing that 
throughout this region the threshold values correspond to equal brightnesses as far as 
rod vision is concerned. For observer A. the curve rises steadily from the yellow to 
the violet. This rise is probably due to a slight colouration of the lens, which is 
indicated also by other measurements. Owing to this colouration the values assumed 
for the luminosity are too high for A. In the case of An. we seem to have a curve 
intermediate in character between those corresponding to the two classes. It has a 
well-marked maximum at SSN 45. 
At 1‘25 degrees from the fovea the results for the observers of class I. obtained by 
using the rod luminosity curve approximate in the case of three observers to those of 
