LOCUS OF COLOR VISION 523 



'light-cells' and 'color-cells', and believed them to have ultimate connec- 

 tions respectively with the rods and cones of the retina. Color cells were 

 especially numerous in Nemestrinus, which has a fovea and has been 

 shown to have color vision. They were sparse in Lemur macaco, which, 

 though diurnal and provided with an area centralis (if not a fovea), 

 probably has no more color vision than L. mongoz (v.s.). Color cells 

 were entirely lacking in the pure-rod Perodict'tcus potto, which Henschen 

 consequently suggested would prove to be the only primate, among those 

 examined by him, entirely devoid of a color sense. 



These investigations have never been carried further; but it would 

 be most interesting to compare, for example, the layer IV's of diurnal 

 squirrels and flying-squirrels — one might find that the two types of 

 cortical cells represented rods and cones right enough, but not neces- 

 sarily achromatic versus chromatic sensory capacity. More interesting 

 still would be someone's demonstration of an analogous histological 

 duplicity in the visual centers of some of the many sub-mammalian 

 possessors of duplex retinae, known either to have color vision, or not 

 to have it. 



(B) Dermal Color-Changes 



No class of vertebrates is lacking in members which, from time to time, 

 alter their color patterns by some means or other. There are vast differ- 

 ences from group to group as to the means employed, the length of time 

 involved, the facility and frequency of the changes, and their biological 

 values. The basic color patterns themselves, and those of animals which 

 cannot change them at all, may or may not be demonstrably adaptive in 

 particular cases. The somber colorations of strictly nocturnal mammals 

 are almost certainly not, for they pass unseen anyway. But we like to 

 think that the vertical stripes of a tiger help to hide him in a canebrake. 

 Fishes are dark above and pale beneath, so that they blend with the bot- 

 tom or with the bright water surface depending upon the point of view 

 of the beholder. We feel sure that this pattern is adaptive — and feel con- 

 vinced when we are confronted by such a phenomenon as the African 

 catfish Synodontis, which swims upside down and whose reversed color- 

 ation is expressed by its Arabic name, 'batensoda' i.- 'black belly') . 



To be sure, the theory of warning and protective coloration is in dis- 

 repute as regards any universal applicability; but there remains an un- 

 shakable residuum of evidence that concealing colorations exist and actu- 

 ally do protect. There have even been experimental demonstrations. In 



