174 Mr. H. M. Bernavd on 



tary stages in the development of tlie sense of vision, is capable 

 of actual demonstration, and on this solid foundation the rest 

 is built. The truth of tliis most important " rest," which 

 includes the phenomena of clear vision such as we know it, 

 can only perhaps be finally established by degrees. Facts 

 are, however, not wanting which make me believe that a 

 rigid demonstration of it is not far off. 



I will point first of all to those observations which tend to 

 show that, in cases of frogs killed after their eyes had been 

 exposed to different coloured lights, the pigment is found 

 massed in a manner not unlike that in which, according to 

 this theory, it should be massed. A chromatic scale, some- 

 what like that whicli I have sketched theoretically, is actually 

 claimed to exist. The red light is found to mass the pigment 

 round the tips of the rods, and so on in regular order upward. 

 It is hardly to be expected that the records of actual dis- 

 coveries in connexion with this chromatic scale would tally 

 exactly with the requirements of the theory. The extreme 

 mobility of the pigmented grarmles would render their per- 

 sistence in any position, after the conditions which induced it 

 were changed, highly improbable. It is, in fact, a matter of 

 surprise to me that they show as much of the theoretical 

 chromatic scale as they are said to do *. 



Again, evidence in favour of the theory, to some minds 

 perhaps of even greater weight than tiiat already adduced, 

 may be found in the fact that it enables us to explain such 

 collateral phenomena as irradiation, contrasts, after-images. 



Irradiation would be due to the increase in size of any 

 bundle of strongly illuminated rods owing to the great crush 

 and continued pressure of granules into the gangways between 

 them. This inci'ease in size of the area occupied by the 

 illuminated rods would press upon the adjacent rods all round 

 for a short distance, the pressure being soon neutralized by 

 the emptying of the gangways (by the squeezing out of the 

 granules) between these adjacent rods. 



On the withdrawal of the light the crush in the gangways, 

 if it has been very great owing to the brilliancy of the illu- 

 mination, takes some time to relieve, during wliich time we 

 have a positive after-image. As soon as it is sufficiently 

 relieved, the granules which had been forcibly squeezed out 

 from between the adjoining gangways force their way back 

 again, and in doing so seem to assist in squeezing the recently 

 congested passages empty. The positive after-image then 

 changes, as in a moment, into a negative after-image with a 



* Cf. Angellufci, Mollescbott's ' Untersiichungen,' xiv. 1890, p. 231. 



