484 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1918. 



sea urchin, or any other such titbit, carnivorous fishes of small or 

 medium size may be drawn from point to point, and their color changes 

 evoked at will. In the face of these and other facts even the most 

 skeptical must accept the complex mechanism of color change in 

 fishes, as a device whose chief function is to enable the species that 

 possess it to display obliterative hues in typical surroundings within 

 their specific ranges, in which they are and have been through ages 

 accustomed to move in the course of their normal activities. 



It is difficult and perhaps impossible to translate the details of a 

 fish's coloration into terms of service rendered. It is conceivable 

 and highly probable that many of its lesser peculiarities are without 

 biological significance. It is clear, however, that in the gross it serves 

 to blend its possessor with its environment and tends to obliterate it, 

 one should suppose, in the eyes of potential enemies or prospective 

 prey. In view of the incidence of the power of color change within 

 the group, and of the correlation of the color of fishes with their habits 

 and distribution, there is no reason to suppose that revealing and 

 concealing types of coloration may be distinguished among them. 

 Broadly conceived, all their coloration, so far as the evidence goes, 

 is obliterative in effect. 



This fact is rich in suggestion. All the Darwinian hypotheses of 

 animal coloration have really grown out of effort to explain how 

 animals' colors and patterns serve them in the struggle for existence. 

 For if such characteristic features as the markings of animals are 

 without utility, the hypothesis of organic evolution by natural 

 selection possesses at best no general application. This is indeed 

 an opinion held by many biologists. But let Darwinians, in a 

 day when the lives of men by thousands have been staked upon 

 the truth of a contrary conception, forswear allegiance to the 

 fetish of warning coloration. Let them spend in the field what- 

 ever time is necessary in order to determine in what relation 

 the colors of birds and insects stand to those of their surroundings, 

 and whether they are correlated with their significant habits and dis- 

 tribution. Then we may gain at last a comprehensive view of animal 

 coloration consistent with fact. In that case, if we may hazard a 

 prophecy, the hypothesis of natural selection should regain in time 

 much of its departed glory; for there is reason to believe that the 

 doctrine of utility will be generally sustained in regard to the colora- 

 tion of the higher animals. 



The prospective significance of known facts regarding the colora- 

 tion of reef fishes is not readily to be stated. It is largely bound 

 up with the subject of changeable patterns; for not only do fishes 

 change their color and shade, but some have two or more alternative 

 systems of markings in which their colors may appear. 



