THE COLORATION OF REEF FISHES 583 



addition to the present group, it is probably impossible to name 

 as many green fishes as are mentioned here, though the entire 

 fish fauna of the Tortugas be drawn upon. Since this includes 

 more than two hundred species (Jordan and Thompson, '04), 

 the statement above is highly significant. 



There is no reason to believe that all typical fishes of the grass 

 flats are green. However, one fair inference may be drawn from 

 the facts presented, i.e., the more nearly a species is restricted 

 to these places at a given stage of its development, the more 

 probable it is that its pigments include that color. It is appar- 

 ent, therefore, that there is intimate correlation between the 

 colors and habits of tropical reef-fishes. It is scarcely necessary 

 to repeat that this fact, which has been demonstrated in the 

 individual cases of red, gray, the lighter blues and green, is of 

 the utmost importance in its bearing upon the problematic 

 significance of animal coloration. 



The working hypothesis enunciated in the introduction seems 

 adequately sustained. Much more has been done, indeed, 

 than test it by applying it to the reef fishes, but positive results 

 in this phase of the investigation support the suggestion of the 

 obliterative effect of color and pattern, which flows from the all 

 but universal existence of countershading and the common 

 occurrence of adaptive color changes within the group. 



DISCUSSION OF RESULTS 



Fishes which range by day amid the most varied surroundings 

 display the greatest variety of colors. They may also possess the 

 best developed means of defence. In this event there would be 

 secondary correlation of vi\dd color combinations with armament 

 and 'distastefulness.' Whether this is so must be determined 

 before appraising the fact above as a confirmation and extension 

 of the partial truth, or a demonstration of error existing in the 

 color hypotheses that postulate conspicuousness. 



In the diversity and richness of their coloration no large group 

 of fishes surpasses the Labrids and Scarids. These are, there- 

 fore, fit subjects for the present inquiry. If their distasteful- 



