CHAPTER XXIX 



Neural activity in the retina 



RAGNAR GRANIT 1 Nnhel hulitute for Neurophysiology, KaroUnska Instituli-t, Stockholm, Sweden 



CHAPTER C: O N T E N T S 



Outline of Retinal Histology 



Electroretinogram (ERG) 



Neural Patterns 



Stimulus Correlates 



Centrifugal Control 



ERG of Man: Its Clinical Use 



OUTLINE OF RETINAL HISTOLOGY 



THE RETINA consists of a surface layer of receptors, 

 the rods and cones (fig. i), joined to a nervous center 

 which delivers an organized message in terms of im- 

 pulses through the optic nerve. The great works of 

 Ramon y Cajal (123, 124) and Polyak (122) should 

 be consulted for details. The anatomy of the eye and 

 retina throughout the vertebrates has been ably 

 discussed by Walls (146). There is also a recent brief 

 summary by Willmer (147). 



In lower vertebrates it is not always easy to dis- 

 tinguish rods from cones (38, 146). In mammals rods 

 end in knobs and cones in dendrites, but in frogs both 

 types of receptor have dendritic terminals. Rods are 

 generally more elongated and slender than cones but 

 this criterion breaks down in some lizards and birds 

 and in the fovea of the primates in which the elongated 

 cones look like rods. Walls emphasizes that the outer 

 cone segment is enclosed by a tubular process from 

 the pigment epithelium cell oppo.site to it and holds 

 this criterion to be universal and never found in rods. 

 Differentiation between rods and cones seems pos- 

 sible by electron microscopy, at least in some species 

 (132, 133, 134, 135). It has even been possible to 

 distinguish two kinds of rods in guinea pigs (135) 

 which would agree well with the electrophysiological 



observations on blue sensiti\ity existing in this spe- 

 cies which has almost no cones. Photodichroism, an 

 orientation of the light-absorbing molecules serving 

 to aid absorption, has been observed in the rods (36, 

 1 29). The fresh cones, viewed end on, light up when 

 the micro.scope is focused on the outer limbs which 

 thus seem to serve as a focusing device operating by 

 total internal reflection (138). This observation may 

 explain why a pencil of light entering at an angle is 

 dimmed if it enters cones, the Stiles-Crawford effect 



(136). 



Double cones and twin cones have been described 

 in fish but since these types do not occur in mammals, 

 they have attracted little attention, physiological work 

 rather tending to settle on universal characteristics. 

 Schwalbes green' rods found in frogs have recently 

 been observed to contain a special blue-absorbing 

 photosensitive substance (37) and ma\' well be more 

 general than one had thought. 



The rods are integrating organs and converge in 

 large numbers towards the bipolar dendrites. Bipolars 

 in their turn converge towards the ganglion cells 

 which give rise to the optic nerve fibers. In man there 

 are some 125,000,000 rods as against 800,000 to 1,000,- 

 000 optic nerve fibers. Since man has only 4 to 7 mil- 

 lion cones, it is clear that the amount of convergence 

 is far less for them; this criterion seems to be general 

 throughout the animal kingdom, signifying that 

 cones, on the whole, are more discriminative, while 

 rods are more integrative and designed to serve as 

 collectors of light quanta in the dark (92, 127). The 

 fundamental observation that rods actually do dom- 

 inate the eyes of nocturnal animals and that an in- 

 creasing number of cones is characteristic of diurnal 

 habits was made by Schultze (131) on the basis of 

 extensive histological studies. This, and later psycho- 

 physical work by Parinaud (117), Konig (99) and 



693 



