Principles underlying the Flicker Photometer. 69 



an equality of brightness one, the author does not mean to 

 convey that it is impossible to demonstrate the profound 

 change in retinal conditions responsible for this effect by a 

 flicker method. 



On the above supposition, it is only the struggle for pre- 

 dominance between the rods and cones that seems to be 

 affected, owing to the difficulty of finding a speed to suit both 

 types of organs. But there seems no doubt that, if the speed 

 is sufficiently slow, a variety of flicker can exist at an order 

 of illumination when we should be justified, apparently, in 

 concluding that the rods are the main instruments of vision. 

 Naturally, therefore, any flicker phenomena produced at such 

 illuminations must be connected with the peculiarities of 

 rod-vision, and may therefore even be used to show the 

 existence of the Purkinje effect. For instance, it is easy to 

 show that a pure red surface appears black and gives no 

 flicker under such conditions. Haycraft, indeed (Proc. Roy. 

 Soc. Lond. vol. lxi. 1897), has even used a purely flicker 

 method to demonstrate the difference in the spectrum-curve 

 of luminosity obtained at high and low illuminations, and 

 thus, in a sense, to exhibit the Purkinje effect as this term is 

 sometimes understood. But this result was naturally to be 

 expected, and does not seem inconsistent with the suggestion 

 the author has sought to convey. 



The Theory of Rod-flicker and Cone-flicker. 



It has been suggested that the means bv which vision is 

 effected through the rods may differ essentially from those 

 utilized in the cone-organs, and it would therefore not be sur- 

 prising to find that there was a distinct difference in the time 

 for which a luminous impression could be retained in the two 

 cases. Thus the visual purple which occurs in the rods, but 

 not in the cones, has long been supposed to be the basis of a 

 photochemical process by which the former organs achieve 

 vision. Since no such substance seems to have yet been isolated 

 in the cones (though Edridge Green has recently pointed out 

 that the visual purple seems to exist between them)*, it would 

 not be unreasonable to suppose that they retained images for a 

 different length of time. 



Let us therefore term the sensation received by using the 

 portion of the retina where rods predominate " rod-flicker,'' 

 and that received through using the part of the retina where 

 cones predominate " cone-flicker." 



A very interesting confirmation of the suggestion put 



* ' Illuminating Engineer ' (London), March 1909, p. 2l0. 





