6 



HOW WE SEE 



by appropriate standardizing action by 

 visual scientists. It should be noted that 

 among the units used in the present chapters, 

 the foot-lambert and the milli-lambert are 

 essentially equivalent, and that the mile 

 candle and the micro-microlambert are 

 convenient for use in special studies because 

 of their small size. 



Photometry. The most commonly used 

 technique in photometric measurement con- 

 sists of comparing the light to be evaluated 

 with some standard, controllable source, 

 using the observer's eye as a null indicator. 

 This technique is perfectly satisfactory for 

 measurements at high and relatively high 

 luminances, but at low luminances it has 

 limitations, particularly when the luminous 

 flux is restricted in its spectral composition. 

 These limitations arise because of the 

 properties of the particular visual cells 

 used in making the photometric comparison. 

 Problems of low luminance photometry 

 will arise in a number of places throughout 

 this and other chapters. They deserve 

 investigation in and of themselves, for 

 unless photometric techniques are stand- 

 ardized, research results at low luminance 

 levels will always be difficult to compare. 

 For a further and more systematic discussion 

 of this stumbling-block to sound, low 

 luminance visual studies, the reader is again 

 referred to the appendix to this chapter. 



Dark Adaptation and Night Vision 



From the standpoint of its function, the 

 eye may be considered as two separate 

 systems. One system operates mainly at 

 ordinary illuminances, such as those en- 

 countered throughout the day and in 

 nonnally lighted rooms at night. The 

 other system takes over at very dim illumi- 

 nances, under starlight and moonlight 

 conditions. There are some important 

 differences in the way seeing is accomplished 

 when each system is in operation and, 

 during the last war, many articles and 

 pamphlets were written on the subject of 

 how to see at night. In the main, these 



publications were essentially correct — as far 

 as they went. They erred chiefly in 

 omitting many important experimental find- 

 ings. These errors of omission were ex- 

 cusable at the time they were committed, 

 but unfortunately they tend to be per- 

 petuated and foster many incorrect, or only 

 partially correct, ideas about dark adapta- 

 tion and night vision. The following 

 discussion, it is hoped, will rectify this 

 situation to some extent. 



Cone and Rod Vision 



Histological studies have provided us with 

 an anatomical explanation for the differences 

 between day and night vision. The back 

 part of the eyeball, the retina, contains two 

 kinds of light-sensitive cells, the cones and 

 rods. Several fines of experimental evidence 

 show fairly conclusively that the cones are 

 the receptors associated with day vision, and 

 the rods those associated with night vision. 

 There is also good reason to believe that 

 individual cones and rods differ in sensitivity 

 and that both types of cells operate over a 

 wide range of illuminance. In general, 

 however, vision in daylight is largely a 

 function of the cones, while the rods are 

 mainly responsible for vision at levels of 

 illuminance below full moonfight. 



Distribution of Cones and Rods in the 

 Retina. Cones and rods are not distributed 

 evenly throughout the retina. The classical 

 data on this point come from the histological 

 examination of a human retina by 0sterberg 

 (72) in 1935. 0sterberg counted the num- 

 ber of cones and rods in areas of a standard 

 size at 164 different positions on the retina. 

 The data plotted in Fig. 2 are for those areas 

 which were sampled on or close to a hori- 

 zontal line through the center of the retina. 

 Although these data are the result of a 

 prodigious amount of labor, it should be 

 understood that they were obtained on a 

 single eye. 



Fig. 2 shows that the density of cones is 

 very much the same throughout most of the 

 retina. In a smafi area around the fovea, 



