Pattern Adaptation of Fishes 81 



patterns (e.g., a chess board), which form the bottom of the 

 aquarium. This reproduction of the pattern is somewhat 

 imperfect, but if we deduct the secondary disturbinj^ fac-tor.s the 

 fact remains that the pattern on the skin is a tolerably true 

 picture of the pattern of the ground. There exists, therefore, a 

 definite arrangement of the images of the (Ufferent luminous 

 points of the ground on the retina and a similar arrangement of 

 the images of the luminous points on the skin of the fishes. We 

 may consider each point of the retinal image as a luminous or a 

 stimulating point which produces a corresponding image point 

 in the primary optic ganglia through the action of th<' nerve- 

 fiber through which it is connected with the ganglia. Every 

 image point in the primary optic ganglia may be considered 

 again as a luminous or stimulating point which through the 

 mediation of a special nerve-fiber influences an individual 

 chromatophore or a small group of chromatophores of the skin. 

 Considering the fact that the retina is a mosaic, we cannot well 

 imagine the transmission of the retinal image uj)on the skin 

 in any other way than by assuming that the relative arrange- 

 ment of the individual points of the retinal image is preserved 

 in the optic fibers and the end ganglia of the optic nerve. 

 Under this assumption a relative distribution of the stimulating 

 intensities must occur in the primary optic ganglion which cor- 

 responds to the distribution of the image points on the retina 

 and which again can be called an image. 



These observations in fish and the conclusions drawn in this 

 note suggest the idea that vision is a kind of telei)hotograpliy. 



