The importance of the mind's eye in ordinary vision is also well illustrated in cases in which we see or seem to see what is not really present, but what for one cause or another it is natural to suppose is present. A very familiar instance of this process is the constant overlooking of misprints--false letters, transposed letters, and missing letters--unless these happen to be particularly striking. We see only the general physiognomy of the word, and the detailed features are supplied from within; in this case it is the expected that happens. In a series of experiments by Professor Münsterberg a word was briefly shown, while just before a certain idea or train of thought was suggested. Under these circumstances the word shown was often misread in accordance with the suggested idea; if the idea of future is suggested, part may be read as past; if vegetable is the suggested line of thought, fright may be read as fruit, and so on. Reading is thus done largely by the mental eye; and entire words, obviously suggested by the context, are sometimes read in, when they have been accidentally omitted. This is more apt to occur with the irregular characters used in manuscript than in the more distinct forms of the printed alphabet, and is particularly frequent in reading over what one has himself written. In reading proof, however, we are eager to detect misprints, and this change in attitude helps to make them visible. It is very difficult to illustrate this process intentionally, because the knowledge that one's powers of observation are about to be tested places one on one's guard, and thus suppresses the natural activity of the mind's eye and draws unusual attention to objective details. Let the reader at this point hold the page at some distance off--say, eight or twelve feet--and draw an exact reproduction of the letters shown in Fig. 2. He should not look at Fig. 2 at close range nor read further in the text until this has been done; and _perhaps_ he may find that he has introduced strokes which were not present in the original. If this is not the case, let him try the test upon those who are ignorant of its nature, and he will find that most persons will supply light lines to complete the contours of the letters, which in the original are suggested but not really present; the original outline, Fig. 2_a_, becomes something like Fig. 2_b_, and so on for the rest of the letters. The physical eye sees the former, but the mental eye sees the latter. I tried this experiment with a class of some thirty University students of Psychology, and, although they were disposed to be quite critical and suspected some kind of an illusion, only three or four drew the letters correctly; all the rest filled in the imaginary light contours; some even drew them as heavily as the real strokes. I followed this by an experiment of a similar character. I placed upon a table a figure (Fig. 3) made of light cardboard, fastened to