Take a large pea or a small marble or bullet and place it on the table or in the palm of the left hand. Then cross the fingers of the right hand as shown in the engraving, Fig. 22, the second finger crossing the first, and place them on the ball, so that the latter may lie between the fingers, as figured in the cut. If the pea or ball be now rolled about, the sensation is apparently that given by two peas under the fingers, and this illusion is so strong that it cannot be dispelled by calling in any of the other senses (the sense of sight for example) as is usually the case under similar circumstances. We may try and try, but it will only be after considerable experience that we shall learn to disregard the apparent impression that there are two balls. The cause of this illusion is readily found. In the ordinary position of the fingers the same ball cannot touch at the same time the exterior sides of two adjoining fingers. When the two fingers are crossed, the conditions are exceptionally changed, but the instinctive interpretation remains the same, unless a frequent repetition of the experiment has overcome the effect of our first education on this point. The experiment, in fact has to be repeated a great number of times to make the illusion become less and less appreciable. But of all the senses, that of sight is the most liable to error and illusion, as the following simple illustrations will show. [Illustration: Fig. 23.] [Illustration: Fig. 24.] In Fig. 23 a black spot has been placed on a white ground, and in Fig. 24 a white spot is placed on a black ground; which is the larger, the black spot or the white one? To every eye the white spot will appear to be the largest, but as a matter of fact they are both the same size. This curious effect is attributed by Helmholtz to what is called irradiation. The eye may also be greatly deceived even in regard to the length of lines placed side by side. Thus, in Fig. 25 a thin vertical line stands upon a thick horizontal one; although the two lines are of precisely the same length, the vertical one seems to be considerably longer than the other. [Illustration: Fig. 25.] In Figs. 26 and 27 a series of vertical and horizontal lines are shown, and in both forms the space that is covered seems to be longer one way than the other. As a matter of fact the space in each case is a perfect square, and the apparent difference in width and height depends upon whether the lines are vertical or horizontal. [Illustration: Fig. 26.] [Illustration: Fig. 27.] Advantage is taken of this curious illusion in decorating rooms and in selecting dresses. Stout ladies of taste avoid dress goods having horizontal stripes, and ladies of the opposite conformation avoid those in which the stripes are vertical.