36'it Discoveries of Miiller a?id others in the 



of the general impression of the light. A spherical visual 

 organ, illuminated by rays of different colours coming from 

 numerous and distinct places externally, will present, in the 

 distribution of the light on its spherical surface, a very im- 

 perfect separation of the different colours. One section of 

 the sphere, for example, vf\\\ be illuminated more by the 

 red, another more by the blue rays. Each set of rays will 

 probably fall upon a large part of the eye ; and, even if the 

 mingling of the clear, the shaded, and the coloured rays be 

 not very great, there may, and probably would, be perceived 

 only one intermediate coloured light. 



The condition requisite for distinct vision would be, so to 

 insulate and limit the light given out from the different 

 points of the object viewed, that it may fall upon the sphe- 

 rical retina at certain points corresponding to the points of 

 emission. If a certain point of the retina can only receive 

 the rays emitted from a certain point of the exterior object, 

 whilst the rays from this point are excluded from all the 

 other parts of the retina, an image of the object will be 

 formed upon the sentient surface. This is exactly what 

 takes place in the compound eyes of insects and of the Crus- 

 tacea, by means of the transparent cones situated between 

 the extremities of the optic filaments and the facets of the 

 cornea. Each of these cones, thus placed on the periphery 

 of a convex nervous mass, conveys to the nervous filament, 

 to which its apex corresponds, that light alone whose course 

 coincides directly with the long axis of the cone itself. 

 The rest of the light given out from some point of the 

 exterior object, falling obliquely upon the cornea, does not 

 penetrate to the internal extremity of the cones, and con- 

 sequently produces no impression upon other filaments of 

 the optic nerve; for, entering the cones obliquely, it im- 

 pinges upon, and is absorbed by, the pigment which sur- 

 rounds them. 



Fig. 6Q, represents the section of a compound eye, in order 

 to show the course of the light. If rays of different colours, 

 given out from the points «, h^ c, d^ fall upon the eye, the 

 cone h will be illuminated throughout its whole length by 

 the ray d'y which traverses this cone in the direction of its 

 long axis. The other cones situated in the vicinity of the 

 line m d will not be illuminated as far as their internal 

 extremity by the rays from c?, which will penetrate less and 

 less deeply into the neighbouring cones, in proportion as they 

 become more remote from the line m d. The nervous fila- 

 ment 7w, corresponding to the cone ^, is consequently im- 

 pressed with the ray d' ; other rays from d, being absorbed 



