Physiological Perspective. 313 



If the binocular examination of the stereograph necessitate 

 these contractions, the corresponding judgment which follows 

 is dependent upon the observer's ability to interpret nerve- 

 impressions, muscular as well as retinal, and not on the 

 mathematical relation between his visual lines. For the 

 apparent relation between objects in the binocular field of 

 view, as influenced by the temporary condition of the muscles 

 of the eyes. I have proposed the name of physiological per- 

 spective *, in contrast with what is suggested by the ordinary 

 elements, whose total effect maybe called physical perspective. 

 The judgment suggested by the former may either coincide or 

 conflict with that suggested by the latter. The result is 

 either enhancement, or partial reversion, or total reversion, 

 according to the arrangement of conditions. The convexity 

 of the spectral wall which Brewster observed, and which is 

 inexplicable on his theory of visual triangulation, is an illus- 

 tration of both physiological and physical perspective ; and 

 by varying the conditions reversion is easily produced. In 

 discussing it, any reference to corresponding retinal points 

 must be understood in the modified sense already explained, 

 and not mathematically. 



5. A New Mode of Stereoscope/. 

 Brewster's experiment may be performed more satisfactorily 

 by constructing on cardboard a pair of perfectly similar 

 figures, each consisting of a series of concentric circles. 

 Combining the images of these by bringing the card quite 

 near and crossing the visual lines at an angle of at least 60°, 

 the binocular image appears suspended in mid air, diminished 

 in size, more especially so in a direction parallel to the inter- 

 ocular line, but convex toward the observer, to whom the 

 appearance is that of a miniature elliptic buckler. The ex- 

 periment is quite straining upon the muscles of the eyes. A 

 far easier method is to cut the card across at the middle, and 

 incline the two halves so that at the edges in contact the two 

 faces shall form a dihedral angle of 120°, or less, opening 

 toward the observer. If the interval between the two centres 

 be equal to the observer's interocular distance, or slightly 

 exceed this, he may easily secure binocular combination with 

 the visual lines parallel or divergent; and the convex shield 

 is seen, larger than in the previous experiment, and sensibly 

 a little more distant than when either picture is viewed 

 separately as in ordinary vision. If there be difficult v in 

 relaxing the muscles of the eyes sufficiently, the cards may 

 be properly held in front of the semi-lenses of an ordinary 

 * American Journal of Science, Xov. 1551, p. 361. 



