6. _Disappearance of lights falling upon fovea._--If the cones are not sensitive to light, a ray of light falling upon the fovea alone and not upon the adjacent portion of the retina containing rods should produce no sensation of light, provided that there is not already any visual purple in the fovea. It has been known to astronomers for a long time that if a small star in a dark portion of the sky be steadfastly looked at, it will disappear from view, whilst other stars seen by indirect vision remain conspicuously visible. The following simple experiment shows the same thing. If a piece of black velvet about three feet square on a door have a pin put in the centre, and the source of light be behind the observer, the pin will be brightly illuminated; and on looking at it (the observer not being too close) and keeping the eye quite still, the pin will disappear, the visual substance diffused into the fovea centralis being used up and not renewed. When viewed by indirect vision it is impossible to make it disappear in this way. When I have taken great care to have very dark surroundings and have used only one eye, I have made moderately bright lights disappear in this manner. These facts have been attributed to a defective sensibility of the fovea for feeble light. The important point, however, that the light is at first most clearly seen by the fovea and only subsequently fades, has been overlooked. If these facts were due to a defective sensibility of the fovea, the star or light would not be visible at first. 7. _Illusion of moving light._--If a small light be looked at fixedly in a dark room, it will appear to move until it comes apparently so close that it could be grasped. The reason of this is that the eye moves so that the light falls upon a more peripheral part of the retina. I find that the movement takes place as if some photo-chemical substance acted under the influence of gravity. For instance, when standing the light appears to travel upwards; resting the head on one side, it appears to travel in the opposite direction. The light appears as if we were looking straight at it, and the eye, which is covered up, remains directed straight at the object. When the second eye is opened two images of the light are seen, and the image which is seen with the periphery of the first eye rapidly coalesces with that seen directly by the second eye. 8. _Purple after-image._--A positive after-image of a purple (rose) colour can be obtained after white light or any spectral colour. It will be noticed that when there is little light during the subsequent observation the colour of an after-image inclines to blue or green, when there is more light towards purple or red. [Illustration: Fig. 1.]