466 ON PHYSIOLOGICAL LIMITS OF MICROSCOPIC VISION. 



between them. In using stripes and wires, not only the interspaces, 

 but also the thickness of the stripes or wires come under considera- 

 tion, and in the calculation Helmholtz has, therefore, assumed the 

 angle corresponding to the sum of a line and an interspace — that 

 is,, to the distances of the central points of the two adjoining 

 objects. The retinal elements must then, at least, he less than the 

 retinal images corresponding to this angle. Harting and Bergmann 

 have some measurements in which the angle thus calculated is less 

 than 60 seconds. Almost invariably, however, it amounts to from 

 60 to 90 seconds. By using extremely thin cobweb filaments, the 

 angle in Harting's experiments proved much greater (2 to 3') than 

 when metallic gauze with thicker filaments was employed. To this 

 cause, no doubt, it is also to be attributed that Yolkmann, who 

 made use of cobweb filaments, found particularly high values." 

 (See article in note to page 460). 



The dimensions of the cones of the retina are, as above quoted, 

 -^-^QQ inch for their thickness, and -^q-^ inch for their distance from 

 centre to centre as they stand closely crowded together at the 

 ''yellow spot" on the axis of vision. But outside of this axial 

 spot the diameter of the cones increases up to about :joVo iiich, and 

 towards the equator of the eye are separate from each other by a 

 still wider interspace of -j-gVo ^^ 'a'io's inch. This interspace is 

 entirely occupied by the rod elements, whose thickness is about 

 T23"o^7 inch, and their distance from centre to centre about i^js-qis 

 inch. (See figures 1, 2, 3). 



View of the Surface of tlie Layer of Rods and Cones on which the 

 Retinal Image is formed^ magnified about 500 times. 



Fig. 1. — Cones of tlie yellow spot. 



Fig. 2. — Cones on the outer margin of the yellow spot with intervening rods. 



Fig. 3.— Cones and rods in the equatorial region of the eye. 



Fig. 



