472 ADAPTATIONS TO PHOTIC QUALITY 



are complementary for him, and so on. The animal's threshold of bright- 

 ness-difference will be nice to know — indeed, this value, determined with 

 white lights, might be worth doing first of all, for it will give you an 

 idea as to how much you dare change, at one step, the intensity of a 

 negative stimulus with which you are trying to confuse the animal in 

 your search for hue-discriminatory capacity. 



Now, no one has done all of these things with any vertebrate species 

 other than Homo sapiens himself. Even if all of the really careful work 

 that has been done, on all vertebrates, had been done on some one 

 species, we would not know quite all of these things about that one ani- 

 mal. Far from being able to compare color-vision systems, all we now 

 know positively — from well-conceived experimentation — is that a few 

 animals do see colors and that a few others do not, and that apparently 

 the color-seeing forms all have a mechanism much like our own. Obvi- 

 ously very, very much remains to be done! 



In the following review of the experimental literature, the fishes are 

 dealt with fairly completely — not because their color vision is any more 

 interesting than that of other groups, but because it is better known and 

 more different investigators have applied more different methods, with 

 more different advantages and faults of technique and interpretation, to 

 its study. The procedures and results with higher vertebrates are describ- 

 ed more sketchily, since after having a given procedure once character- 

 ized for him, the reader can be spared any detailed reiterations of favor- 

 able and unfavorable criticisms. 



Fishes — The reader's suspense, if any, may as well be relieved at once 

 by the flat, if somewhat back-handed, statement that no fish is known 

 not to have color vision. But the angler can take little comfort from the 

 fact. As will appear shortly, he cannot predict whether a red fly will 

 attract or violently repel the fish he is after. In fact, there is every reason 

 to think that a dry fly, or a floating plug of any color, is seen by the fish 

 merely as a dark silhouette whose form is much more important than its 

 hue. Much though we may know of the color vision of laboratory fishes, 

 and infer as to the color vision of game species, when it comes to wet 

 flies and plugs the old rule still holds : what they'll take, they take, and 

 what they won't, they don't. 



The first scientific work on fish color vision was reported by Graber in 

 1884 and 1885. He made use only of untrained responses made by the 

 fish toward different stimuli presented in pairs — the so-called color-pre- 

 ference technique. Working with the freshwater Bar ba tula barbatula and 



