Passing to the fifth and last variety of irradiation, this seems due to fatigue in the inhibiting apparatus which reduced the spread of the first kind of irradiation. Following out the scheme we have applied, it would seem as if the channels which were first opened by the direct stimulation became blocked through fatigue, and, therefore, the excitation produced in the retina were forced to seek new paths through to the brain by means of the nerves which proceed from the unstimulated portions of the retina. Thus if the resistance through fatigue occurs slowly, the excitation which spreads may increase in intensity and in extent. So, as the resistance increased, a portion corresponding to the directly stimulated portion and its slight irradiatory enlargement of the borders would be surrounded by a cloud of light growing in size and intensity. Of course the limiting case would be when the external cloud of light attained as great or even greater intensity as the stimulated portion, but such a case would probably be impossible to realize because of other conditions which would prevent. It may be that this fifth variety is caused partly at least by a cortical spreading of the excitation. But it seems to me more likely, in view of the fact that we could find no irradiatory enlargement of the binocular portions of stereoscopic images and for a number of other reasons, that the induction is retinal in character and that after the resistance through fatigue has arisen in the central organs the stimulation spreads out over the retina to the unstimulated portions of the field and proceeds from thence to the brain. This seems more probable than that the stimulation continues to be confined merely to the stimulated portion of the retina, but seeks passage from one portion to another of the brain through fresh neurones which branch only from those nerve-tracts which proceed from the directly stimulated portions of the retina. To conclude; we have seen that there are various forms of irradiation which take place during the perception of stationary and moving sources of illumination. That there are certain modifications in the form of a moving image which are probably due to one of these processes. Concerning color irradiation it was found that the curvature of the images varied with the color of the light, so that a figure illuminated by a colored light of one intensity would not have the same curvature as one illuminated by a light of the same intensity but of another color. Green gives the greatest curvature, yellow the next, red and blue about the same. In other words the differences in curvature of the images follow the order of the brightness of the colors in a spectrum the intensity of which is much reduced.