A very remarkable illusion, affecting the apparent position of the drawings of geometrical solids, was first observed by the late Professor Neckar, of Geneva, who communicated it to me personally in 1832.[73] “The rhomboid AX,” (Fig. 51,) he says, “is drawn so that the solid angle A should be seen nearest to the spectator, and the solid angle X the farthest from him, and that the face ACBD should be the foremost, while the face XDC is behind. But in looking repeatedly at the same figure, you will perceive that at times the apparent position of the rhomboid is so changed that the solid angle X will appear the nearest, and the solid angle A the farthest, and that the face ACDB will recede behind the face XDC, which will come forward,—which effect gives to the whole solid a quite contrary apparent inclination.” Professor Neckar observed this change “as well with one as with both eyes,” and he considered it as owing “to an involuntary change in the adjustment of the eye for obtaining distinct vision. And that whenever the point of distinct vision on the retina was directed to the angle A for instance, this angle, seen more distinctly than the other, was naturally supposed to be nearer and foremost, while the other angles, seen indistinctly, were supposed to be farther away and behind. The reverse took place when the point of distinct vision was brought to bear upon the angle X. What I have said of the solid angles (A and X) is equally true of the edges, those edges upon which the axis of the eye, or the central hole of the retina, are directed, will always appear forward; so that now it seems to me certain that this little, at first so puzzling, phenomenon depends upon the law of distinct vision.” [73] See _Edinburgh Philosophical Journal_, November 1832, vol. i. p. 334. In consequence of completely misunderstanding Mr. Neckar’s explanation of this illusion, Mr. Wheatstone has pronounced it to be erroneous, but there can be no doubt of its correctness; and there are various experiments by which the principle may be illustrated. By hiding with the finger _one_ of the solid angles, or making it indistinct, by a piece of dimmed glass, or throwing a slight shadow over it, the _other_ will appear foremost till the obscuring cause is removed. The experiment may be still more satisfactorily made by holding above the rhomboid a piece of finely ground-glass, the ground side being farthest from the eye, and bringing one edge of it gradually down till it touches the point A, the other edge being kept at a distance from the paper. In this way all the lines diverging from A will become dimmer as they recede from A, and consequently A will appear the most forward point. A similar result will be obtained by putting a black spot upon A, which will have the effect of drawing our attention to A rather than to X.