f,r>o PHYSIOLOGY 



length of time in order to make the greatest impression possible. 

 For coloured lights approximately the same values are found, pro- 

 vided that allowance is made for the comparative intensity of the colour. 

 Since the time required for the retina to reach a steady state is 

 nearly that at which a series of stimuli must fall on the retina in order to 

 produce a uniform sensation, for intensities near the threshold 

 the rate at which flicker disappears is one stimulus every quarter 

 of a second. But it is found that as the intensity rises the rate must be 

 increased in order to abolish flicker. The rule which most nearly expresses 

 the relation appears to be that geometrical increase in the intensity requires 

 an arithmetic increase in the rate. Sherrington showed however that the 

 results are affected by simultaneous contrast. This phenomenon of flicker 

 in, as I have said, used in practice for measuring the intensities of light sources. 

 Two methods are employed, firstly that in which the two light sources to be 

 compared are measured separately for the intensity at which flicker ceases 

 when the same rate of stimulation is used for both; and secondly that in which 

 the two sources are. caused to fall alternately on the eye, and are adjusted 

 in intensity until flicker ceases. Of the tw T o methods the latter is 

 the more accurate. The value of these methods lies in the fact that 

 they measure brightness independently of colour. The shape of the curves 

 obtained by plotting the luminosity of different parts of the spectrum has 

 been shown in Fig. 286. 



Lastly we have to consider the relationship between time intensity and 

 apparent brightness in the case of an intermittent source, the rate of which 

 is sufficiently great to avoid flicker. Experiment shows that the brightness 

 increases in proportion as either the intensity or the time is increased, 

 and further that equal brightness is obtained if the time multiplied by the 

 intensity is constant. This statement is true of both the cone and the rod 

 apparatus, and is known as the Talbot Plateau Law. Use is made of this 

 law in the sector method of controlling intensity (see page 564) because the 

 intensity is proportional to the time during which the light is allowed to 

 pass through, which is in its turn controlled by the angle between the 

 blades of the sector. 



COLOUR THRESHOLD 



On testing the violet end with a photographic process, or the red md 

 by a thermopile, it can readily be shown that the spectrum extrmU ,it 

 both ends far beyond the visible limit. The visible limit at the red 

 end under the most favourable conditions has been found to be 8350 A.U., 

 while under ordinary circumstances it is difficult to go beyond 8000 A.U. 

 Since rays beyond this reach the retina in considerable amount, the limit 

 cannot be caused by opacity of the eye media, and must therefore be due 

 to an actual inability on the part of the retina to record their presence. 



Of several hypotheses which might be advanced for this inability, the 

 most probable is that the retinal pigments are unable to absorb rays in the 

 infra-red part of the spectrum, and therefore according to Draper's lav/ such 



