THE THRESHOLD OF VISION FOR DIFFERENT COLOURED LIGHTS. 
125 
illuminates the disc, allows of the intensity of the illumination being adjusted. By 
removing H and placing a rod so as to form a Bumford photometer, the intensity of 
the beam which goes straight through and is doubly reflected, respectively, can be 
compared. The experiment consists in adjusting the sectors so that when the eyes 
are kept fixed on the centre of the disc the disc and the annulus appear equally 
bright. Settings were taken alternately starting with the centre too bright and 
then too feeble. Observations were made throughout the spectrum with different 
intensities.* 
Two series of measurements were made, one with the eye thoroughly dark adapted, 
the annulus being surrounded with black velvet and all stray light screened off. In 
the other set the eye was light adapted, a sheet of white blotting paper surrounding 
the annulus, and was illuminated by the white light of the arc. A screen was so 
placed that none of this white light fell on the annulus or disc. The intensity of the 
white light on the background was 1'5 lux, and was adjusted so that when a red and 
a blue light were compared by a person who has rods at the fovea the relative 
intensity of the two lights did not appear to alter, as the intensity of both was 
reduced in the same proportion, that is till the so-called Purkinje effect vanished. 
A series of measurements were first made using a constant slit width throughout 
the spectrum by both W. and B., and the results are given in Table VI. and 
Table VI. 
Colour. 
SSN. 
Ratio of central to parafoveal 
illumination for equal brightness. 
Dark adapted eye. 
W. 
B. 
58-7 
658 
1-14 
1-05 
56-0 
632 
1-17 
53-3 
609 
1-23 
1-06 
50-8 
589 
1-29 
48-6 
576 
1-15 
1-16 
45-4 
556 
1-32 
42-8 
541 
1-38 
1-10 
40-1 
527 
1-54 
37-6 
514 
1-70 
1-15 
35-0 
503 
2-42 
32-2 
493 
4 63 
1-12 
27-0 
474 
9-53 
1 -32 
21-8 
457 
• 13-6 
1-24 
16-5 
442 
19-6 
1-34 
11-2 
427 
23-9 
1-41 
5-9 
415 
27-0 
1-49 
0-7 
403 
27'0 
1*94 
1 
* The slit, A, could be varied in width, the actual intensity when the slit was placed at the D line being 
determined by comparison with a Hefner lamp. 
s 2 
