JOURNAL OF SCIENCE AND ARTS.
ART. L.-On some phenomena of Binocular Vision;* by JOSEPH LECONTE, Prof. of Geol. and Nat. Hist., Univ. of California.
V. Stereoscopic phenomena.
IT is a familiar fact that in stereoscopic pictures, properly mounted, identical points in the foregrounds of the two pictures are always a little nearer together than identical points in the backgrounds. With a pair of compasses we can, by this means, easily test whether or not pictures are properly mounted. It is evident therefore that it requires greater optic convergence to unite the foregrounds than the backgrounds of the two pictIt is also evident that we cannot at the same time and with the same convergence unite all parts of the pictures. When objects in the foreground are united, objects in the background are seen double, the images being homonymous; when objects in the background are united, then objects in the foreground are seen double, the images being heteronymous; when objects in the middle ground are united, then both fore and background are doubled but in different directions. In looking at pictures in a stereoscope, therefore, the eyes range rapidly from fore to background and vice versa, uniting the objects successively, and finding the visual phenomena precisely similar in all respects to natural vision of near and distant objects, instinctively introduces the idea of depth of space. Or even looking steadily at any point, say in the middle ground, the depth of space is still perceived, as in nature under similar circumstances, and for the same reason, viz: that the eye or the mind, instinc* For the preceding articles on this subject, see II, xlvii, 68, 153, and III, 1, 33. AM. JOUR. SCI.-THIRD SERIES, VOL. II, No. 7.-JULY, 1871.