THE THRESHOLD OF VISION FOR DIFFERENT COLOURED LIGHTS 
121 
less bright. When the intensity of the light is reduced to such an extent that to an 
observer of class I. there is no sensation of light in the centre, an observer of class II. 
sees a continuous streak of light, the brightness at the centre being little less than 
that at the sides. If the intensity is reduced to about one-twentieth of that 
required for central extinction for class I., the centre also appears dark for class II., 
but the sides then are very faint, in fact the threshold is almost readied throughout. 
If a red (SSN 50, 632 /ul/u.) spot of light is used, then for both classes the streak of 
light looks brighter at the centre than at the sides, whatever the intensity. 
The explanation of these results is that with the green light, which stimulates 
both the rods and the cones if sufficiently bright, when the intensity is reduced below 
the cone threshold in the case of class I. (where there are no rods at the fovea), no 
sensation is produced in this region. In the periphery, where there are rods, the 
stimulus is sufficient to cause the sensation of light. In the case of class II., since 
there are rods at the fovea, the stimulus even when below the cone threshold is 
sufficient to cause a sensation of light by means of these rods. With a red light, 
since this colour is unable to stimulate the rods, we have only to do with cone 
vision, and the results obtained indicate that the cone sensation is a maximum at 
the fovea for observers of both classes. By means of this apparatus we have tested 
ten persons, and as a result find that eight of them belong to class I. and two to 
class II. 
This method of examining, in which there is no difficulty in obtaining correct 
fixation, may perhaps be of considerable assistance to ophthalmologists when examining 
a central scotoma. 
It is rather curious that when one of us (A.) made the investigation as to the 
sensitiveness of the retina described in a previous communication (‘ Phil. Trans.,' 
A, vol. 190, p. 155, 1897) he confirmed his results by means of those obtained by B., 
and as they agreed, he concluded that the results obtained applied to all normal 
eyes. We now know that both A. and B. belong to a class which seems to form only 
a small proportion of normal eyes. 
There has been some considerable discussion whether the secondary image or 
Bidwell’s ghost which is seen to follow a moving spot of light when the eye is kept 
stationary occurs at the fovea. Using the arrangement described above, but causing 
the spot of light to travel much more closely, we have examined this question in the 
case of a typical observer of each class. Using a green stimulus light the ghost is 
very well marked, and to W., an observer of the first class, the ghost always appears 
to jump the fovea. In the case of an observer of the second class, B., the ghost is 
seen to follow the primary right across the field, no interruption at the fovea taking- 
place. It would thus appear that the rods and not the cones are involved in the 
production of this secondary image. * 
* [When this was in print we were not aware of McDougal’s experiments on the Bidwell ghost. His 
paper should be examined in connection with the results given in the above paragraph .—October 20,1915 .] 
