THE VERTEBRATE EYE 261 



the same component parts, the outer segment being typically (but not 

 invariably) conical, the inner segment typically fatter, shorter and 

 more squat than the corresponding part of the rod, often with an oil- 

 droplet in the ellipsoid, and sometimes with a paraboloid composed 

 of glycogen lying more proximally ; the nucleus is relatively larger 

 and the foot-piece more widely spread. Variations to this standard 

 structure are common, such as the presence or absence of oil-droplets, 

 the occurrence of double, triple or even quadruple elements, and so on ; 

 these will be discussed in the sections on systematic anatomy. 



While these are the typical structural features, however, the 

 variations in the morphology of the rods and cones are so marked as to 

 have led to much confusion and some controversy ; some rods resemble 

 cones more closely than some members of their owii family, while the 

 cones of a well-developed fovea often resemble elongated rods more 

 closely than typical cones (Fig. 267c). Indeed, in our systematic 

 survey we shall on more than one occasion run up against difficulties 

 in describing particular visual cells either as a rod or a cone. 



Schultze (1866), who first clearly differentiated the two types of cell, did so 

 primarily on anatomical grounds, his three criteria being — (o) the cylindrical 

 termination of the rods in contrast to the conical tip of the cones, (b) the more 

 external position of the cone-nuclei close to the limiting membrane owing to 

 the shortness of the inner segment, and (c) the knob-like ending of the rods in 

 contrast to the spread-out foot-piece of the cones. Unfortunately, all these 

 conditions ai'e violated, sometimes even in the same retina. The tip of the cone 

 may be slender, elongated and cylindrical (as in lizards and birds, Verrier, 1935 ; 

 Detwiler, 1943) ; the nuclei may lie in a single layer (amphibians, Saxen, 1953), 

 or the usual arrangement may be rev^ersed (some fishes and amphibians : Cajal, 

 1893 ; Franz, 1913 ; Memier, 1929) ; while the foot-pieces of rods may be 

 branched (some fishes, amphibians and birds: Greeff, 1900; Putter, 1912; 

 Detwiler, 1943). 



Differential methods of staining have been attempted as a criterion (Dogiel, 

 1888 ; Kolmer, 1936 ; Wolff, 1949 ; Wislocki and Sidman, 1952 ; Saxen, 1953 ; 

 and others) and again have led to inconclusive results. A further point of 

 differentiation is a study of the connections of the visual cells ; several rods are 

 typically associated with one bipolar cell, while each foveal cone is ordinarily 

 connected with one bijoolar cell ; but again, this relationship is not maintained 

 by the peripheral cones nor in retinae without a fovea. It is possible that, when 

 more fully developed, the study of the ultra-microscopic structure may provide 

 further evidence whereon a distinction between the two types of cell may be based. 



The difficulty in anatomical differentiation has naturally stimulated 

 attempts at a functional basis for classification, for it is generally 

 conceded that the cones mediate photopic (and colour) vision while 

 the rods are concerned with scotopic vision. Tlie physiological distinc- 

 tion between " photocytes " and " scotocytes,"" however, is equally 

 fraught with difficulties. The presence of rhodopsin or its relatives 

 would theoretically substantiate ihe presence of rods, but while this is 



