310 Mr C. V. Burton, Experiments on Colowr-Perception ; [Nov. 26, 
resulting impression may be analytically assigned to white light. 
The present observations seem to indicate that the curves for 
the two extreme sensations overlap in the middle of the spectrum, 
so that all three sensations participate in the perception of colours 
from the yellowish-green to the greenish-blue, whilst not one 
of the three sensations is sensitive throughout the whole range of 
the visible spectrum. 
One or two matters of detail should here be mentioned. It 
was always found that for a considerable time after the right 
eye had been fatigued for a given colour, there was a marked 
difference in the appearance of surrounding objects as seen by 
the two eyes, whilst on closing both eyes, no colours were seen, 
showing that the change effected by fatigue was merely one 
of sensitiveness, and did not involve any persistence of active 
disturbance. Before tiring the eye with another colour, an interval 
of some hours was always allowed to elapse, so that vision had 
again become normal. 
Whilst looking at the unclouded sun through red glass there 
was a gradual alteration in the appearance of the sky, which 
seemed to change from a deep brick-red to a dull foggy-yellow. 
Similar but less marked effects were obtained while looking 
through green or blue glass. No doubt these changes of timt 
are accounted for by the fact that the colour-sensation chiefly 
involved becomes much more completely and rapidly fatigued than 
that which is only moderately excited. 
On looking for some time at the violet end of the spectrum, 
it appears redder to the tired eye. Again spectrum-violet diluted 
with white can be exactly matched by” a mixture of blue and ex- 
treme red. ‘Two possible explanations present themselves. 
(1) That the apparatus which is chiefly sensitive to red im- 
pressions attains a secondary maximum of excitability in the 
neighbourhood of the violet. 
(2) That violet light is partly changed to red by fluorescence 
of the retina, and so affects the red colour-sensation. 
The latter view was experimentally tested in the following 
way. A beam of sunlight passed through a hole in a shutter 
which was covered with a double thickness of cobalt glass. This 
beam was received on the concave mirror of an ophthalmoscope, 
and being concentrated on the eye of the subject, the orange- 
red retina could be distinctly seen, as in the ordinary use of 
the instrument with white hight. The appearance was so bright 
and the hue so orange a red as to leave no doubt that the effect 
was mainly due to fluorescence. The double thickness of cobalt 
glass transmits none of the less refrangible rays, except a dim and 
narrow band in the extreme red. 
Fig. (4) roughly indicates the general nature of the results 
obtained ; the dotted line corresponding to the excitement of the 
red sensation by fluorescence from violet light. 
