THE COLORATION OF REEF FISHES 591 



second class more able to change its color than the green one 

 which is assigned to the first, and the complexity of its pattern 

 is distinctly less than that of Sparisoma viride, if the reference 

 is intended for that species. Hence upon his own testimony 

 it would appear that Bristol dealt with a series in which there are 

 no natural breaks. 



The ease with which observers become convinced of the con- 

 spicuousness of some animals seems to rest upon a twofold error 

 which may be illustrated by reference to fishes, for the presence 

 of red and blue species in surroundings where these colors do not 

 commonly appear, and the occurrence of forms of different color 

 side by side may be justly held responsible for much of the per- 

 sistent misunderstanding that has prevailed. In the first case, 

 seeing nothing in the background the repetition of whose tones 

 one might suppose to minister to the inconspicuousness of the 

 creatures upon which they appear, one is led to believe that there 

 is nothing, and might even infer that the colors of the animals 

 are out of keeping vdih. their surroundings and, therefore, con- 

 spicuous. In the second instance one is tempted to conclude 

 that color is not of great importance, since all sorts of fishes seem 

 to mingle freely at one station. 



There is a natural fish-trap, or weir, formed by a hook at the 

 tip of a small promontory beside the eastern entrance of the 

 channel cutting off San Juan from the mainland of Porto Rico. 

 Atthis place, in an area measuring perhaps 50 by 75 feet in 

 extent, there were counted in the few minutes it required to wade 

 about and examine it thirty-six species of fishes, including all 

 types but the red. In viewing such a motley assemblage, since 

 many of the shades noted obviously fail to repeat local coloring, 

 one who does not realize the unnaturalness of the situation can 

 scarcely be blamed for concluding that many of the hues are 

 either functionally conspicuous, at at least without adaptive 

 significance. 



The same condition prevails in less marked degree on every 

 coral reef under circumstances concerning which no plea of 

 abnormality may be entertained. But one gradually learns 

 that first impressions are unreliable. The wealth of species and 



