THE STRUCTURE OF VISUAL CELLS 39 



the above listed functions to see and be convinced that rod 

 function and cone function are two separate and discrete 

 processes. It is evident, therefore, that we need not refer 

 to the Duphcity Theory as theory but as fact, and that the 

 retina must be regarded as a dual sense organ. In the light 

 of these facts it becomes apparent that the presence and 

 relative distribution of rods and cones in the same eye, as 

 well as in different vertebrate eyes, becomes a matter of 

 first importance to anyone who desires to study vision from a 

 comparative point of view. 



Although rods and cones each have specific characteristics, 

 they are constructed upon a common general plan. Both 

 possess an inner and an outer segment, a nucleus, and a 

 centripetal fiber which brings them into relation with the 

 bipolar transmitting cells. They extend through the pores 

 of the fenestrated external limiting membrane so that their 

 nuclei and centripetal fibers lie internal to it, whereas the 

 main bulk of the cell lies external (Figure 24). In most 

 animals with dual retinas the cone nuclei lie closer to the 

 external limiting membrane than do the rod nuclei (e.g. 

 mammals and most fishes). In amphibia, however, the 

 condition is reversed, as it is also for the dual retinas of 

 reptiles, and presumably for birds, although in this latter 

 group the nuclei of the two types of visual cells are less dis- 

 tinguishable one from the other. 



Cones. Typically the cone is a conical-shaped element 

 possessing an inner and an outer segment. It is the conical 

 shape of the outer segment which is the morphological fea- 

 ture used to distinguish this cell as a cone, whereas the shape 

 of the outer segment of the rod is typically cylindrical. 

 There are marked deviations from the typical condition 

 and cones, as well as rods, vary considerably in form, size, 

 and structure. In the region of the fovea, the cones are long, 

 slender attenuated structures which look more like rods, 

 yet their function fits in with the facts of cone physiology. 

 In the human fovea they are said to measure about 85 ii in 

 length according to Greeff (1899), from 58 to 67 m according 



