194 Mr. S. Rowley on a New Theory of Vision. 



the convergence of the axes, the theory of Smith places them 

 in lines drawn from the eyes to the object somewhere between 

 the place of the object and the point of convergence, " but not 

 very far from the real place," i. e. the place of the object*. 

 Hence, while the two images of the object on either side of the 

 point of convergence will be seen coincident with their object at 

 a distance of 1 T ^- inch from the two images (seen in like man- 

 ner) of the object on the other side, the two images of the re- 

 mote object will be seen at a much less distance from each 

 other, or, following the proportions of the construction given us 

 by that author, at a distance of -fij or A inch. 



Aguilonius is the only author who, meeting the question in 

 its full extent, undertakes to assign for each image both the line 

 of visible direction and the absolute visible distance from the 

 retina on the line of visible direction. 



Sir David Brewster indeed gives both conditions of external 

 visible position for the images falling on the vertices of the retince. 

 But for every other image he merely assigns Kepler's line of 

 visible direction, being silent regarding the visible distance — 

 the essential complement of the former in determining visible 

 position. 



I will now propose and proceed to put to the test of experi- 

 ment what I believe to be the true theory of vision — premising 

 the remark that in establishing its existence it will not render 

 the proof offered less exact if I do not undertake to determine 

 the precise point where a given rayf strikes the retina, nor the 

 precise point at which the line of visible direction crosses the 

 optic axis, provided we admit (what I think will not be disputed) 

 that where the angles with the optic axes of two rays falling one 

 on each cornea are equal, the distances of the images on the re- 

 tinae from the extremities of the axes are equal, and provided we 

 admit (what I also think will not be disputed) that where the 

 angle with the axis of one of the rays is but very slightly in- 

 creased, the angle formed after decussation is not appreciably 

 WTiequally increased — and further premising that I will neglect 

 any effect of refraction, and also suppose a coincidence of the 

 line of the ray with the line of visible direction. 



All the component points of the total impression on the retina of 

 either eye, before becoming cognizable mental phenomena, are si- 

 multaneously, and in lines drawn from them through a point situ- 

 ated, say, a little behind the centre of the crystalline lens, referred 

 outivard to a surface parallel to the impressed surface of the retina 

 conceived to be simply expanded coextensively with the impression 



* Smith's 'Complete System of Optics/ art. 137- 



t Here, and wherever in this paper the word ray is used, it is intended 

 to signify the central or axial ray of the pencil. 



