The fact is that we have to take into consideration two other factors in the process, which we call "seeing," besides the duration of the retinal impression or excitation. These are, first, attention, and second, judgment. We are apt to think that "seeing" is a simple, straightforward sort of thing, whereas it is really a strangely complex and delusive process. "I did not see it, therefore it was not there," or "You must have seen it; it was right in front of you," are common assertions, and the belief that such assertions are justified leads to miscarriage of justice in courts of law. Yet everyone knows that he may stare out of the window of a railway carriage and have a long panorama pass before his eyes, or may walk along a crowded street and look his acquaintances in the face, and in neither case will he have "seen" or recognized anything, or be able to give an account of the scene that was pictured on the back of his eye. Attention, the direction of the mind to the sensation, is necessary; and it appears that it is very difficult (to some more than to others) to hold the attention alert, and to give it to the _unexpected_. In fact, to a very large extent we can only "see" (using the word to signify the ultimate mental condition) that which we are prepared to see or that which we expect to see. In the absence of such expectation, a very strongly illuminated or well-marked, outstanding object is far more readily "seen" than less marked objects. Accordingly, the outstretched legs of the galloping horse, now in front and now behind, are "seen," whilst the rest of the phases are not observed. Moreover, it is a fact that the swinging pendulum of a clock is "seen" at the extreme position of the swing on each side, and not in the intermediate space. This is because the image is formed very quickly, twice in the space where the bob of the pendulum is coming to the limit of its swing and is again returning on its course. For the same reason, the outstretched legs of the horse going up to their limit and at once returning give in very quick succession, near their extreme limit, an ascending and a descending phase which are not strictly but sensibly alike, and so doubly impress the retina, and obtain for the legs "attention" when in that extreme position. The choice of the attitude depicted by Morot is explained by the fact that, as is shown by its persistence through two successive pictures (figs. 2 and 3 of Pl. I), this pose must produce a more continuous impression on the retina than any other of the attitudes shown, since none of them endure through two successive pictures.