IN SINGLE AND BINOCULAR VISION. 351 



pencils ; the visible and the true direction of points would have coincided, and 

 objects would have changed neither their form nor their position during the mo- 

 tion of this hypothetical eyeball round the common centre of the two coats. But 

 as such an eye could not have afforded sufficiently distinct vision, the introduc- 

 tion of the crystalline lens became necessary ; and it is owing to the secondary 

 refractions at its surfaces and within its mass of variable density, that the paral- 

 lax of visible direction is produced. 



The following experiment will establish the existence, and explain the nature 

 of this parallax. Let MN, Fig. 1, be the eyeball, C the centre of curvature of the 

 retina, and also the centre of motion of the eyeball. Having placed an opaque 

 screen S several inches from the eye, till its inner edge just eclipses a luminous ob- 

 ject A, look away from the screen, and the object A will appear. Keeping the head 

 steady, place another screen S'* so that, when viewed directly, it does not eclipse 

 another luminous object B, the line CS'B just grazing the outer edge of B. When 

 the screens and luminous objects, therefore, are so arranged that A is invisible 

 when the axis of the eye is directed to S or to A, and B visible when the axis of 

 the eye is directed to S' or B, then by turning the eye from A to B, A will appear, 

 and B will disappear, exhibiting the curious effect of an invisible body appearing 

 by looking away from it, and of a visible body disappearing by looking at it! 



Had the eyeball MN been our hypothetical one, these effects would not have 

 been produced. All objects, near and remote, would have retained their relative 

 positions and magnitudes during its rotation. 



Hence it follows, that we are not entitled to reject any law of visible direc- 

 tion, because it gives a position to visible objects different from their real posi- 

 tion. 



Having removed this difficulty, I proceeded to examine the other data of 

 D'ALEMBERT. Making the eyeball and the retina spherical, he assumes that the 

 centre of the latter is equidistant from the foramen centrale of the retina, and the 

 centre of the crystalline lens. This, however, is far from being the case. M. Du- 

 TOUE, and Dr THOMAS YOUNG, have made the centre of curvature of the retina 

 coincident with the centre of curvature of the spherical surface of the cornea, as 

 in our hypothetical eye ; and this centre, in place of being almost half-way be- 

 tween the apex of the posterior surface of the lens and the foramen centrale, is 

 actually almost in contact with the latter ! The dissections of Dr KNOX and Mr 

 CLAY WALLACE, of New York, give similar results. When we add to these con- 

 siderations the fact that the refractive power of the crystalline lens assumed by 

 D'ALEMBERT is nearly triple of what it really is, we are entitled to reject the re- 

 sults of his calculations. 



Assuming, then, the most correct anatomy of the eye, namely, that according 



* The two screens S, S' may be the opposite edges fo a triangular notch in a card held in the hand. 



