The fourth kind of irradiation which we have defined is of course the best-known form, and is that which has been the most discussed by the many writers on the subject. It will be remembered that this form appears in stationary objects which have been observed for some little time (from four to ten seconds), and consists in the apparent enlargement of a more intensely illuminated portion at the expense of a less illuminated one. This enlargement occurs after all trace of the first kind of irradiation has vanished, and of course no trace of the third kind comes in, since the object is stationary. The course of events may then be somewhat as follows. In the first perception of the object we have the wide-spreading irradiation described. Then way is made through the synapses corresponding to the stimulated portion of the retina, and the wide-spreading irradiation is drained off through these open channels, so that the image contracts again to its proper size. But at the same time it is not likely that there will not be a slight irradiatory enlargement of the borders of the image. For irradiation is present within the confines of the image. This is shown not only in the case of moving images, but also in the fact that the edges of the less intensely illuminated portions of the field are curved inward, this being most probably due to the fact that the centres of the contiguous luminous portions are reënforced by irradiation proceeding from the direction of both the ends. Not all of the excitation proceeds to the brain from the directly stimulated portions of the image merely, but a little irradiates over the borders and causes an apparent enlargement of the brighter field. It has also been shown by Plateau and others that the amount of irradiation increases both with the intensity of the stimulation and with the time during which it acts. Of course, as to the intensity there is no question. As to the time-element, it may be that the excitation at the border spreads rather slowly outward after the previous contraction of the image to its proper dimensions, which takes place within a very short time after stimulation, until a sort of balance is reached between the tendency of the image to enlarge itself through irradiation and the tendency for this irradiatory excitation to be drained off through the nerves corresponding to the stimulated portion of the retina, after which no further apparent enlargement takes place. In some of our experiments with dots we found that after a dot of the proper intensity of illumination had been steadily gazed at for some time the centre would appear dark. This seems to be due to the fact that the centre of such an image was reënforced by irradiation, so that the nervous mechanism corresponding to it became fatigued more quickly and the stimulation at the centre no longer gave such intense sensations as the rest of the figure, but appeared darker.