88 



THE AMERICAN MUSEUM JOURNAL 



related with spec ific habits. Eed, for 

 example, is almost wholly restricted to 

 nocturnal iishes which lie hidden by 

 day. The greens, on the other hand, 

 characterize such species as live upon the 

 green reef-flats, or swim near the sur- 

 face in open water. The fishes display- 

 ing the most vivid color combinations 

 move in the most varied environments. 

 The colors of all, in brief, appear to 

 conform to a rational and simple sys- 

 tem of distribution. 



The coloration of many of the spe- 

 cies is changeable and may be experi- 

 mentally controlled. Equipped with a 

 diving-hood one may walk about freely 

 and thus lead carnivorous forms from 

 point to point by offering them food. 

 It is possible under such circumstances 

 to evoke their various color phases at 

 will by selecting the spot to which the 

 creatures are led, and, if desired, to 

 observe at closest range each minute 

 effect of their changed environment. 

 The movements of herbivorous forms 

 are not subject to quite such definite 

 control, but one may walk up to brows- 

 ing schools of tangs or parrot fishes, 

 follow them about, and secure special 

 information with little effort. 



All tlie creatures may be photo- 

 graphed amid perfectly natural and 

 typical surroundings with much less 

 difficulty than might be anticipated. 

 Hence it has been possible to make en- 

 couraging progress in an attempt to 

 secure a pictorial record of the phases 

 assumed by a number of species under 

 controlled conditions. 



The two pictures of the hogfish give 

 an idea of the extent to which changes 

 which for all practical purposes are in- 

 stantaneous may modify the appearance 

 of a species. In nature, so far as I have 



been able to determine positively from 

 personal observation, such changes are 

 solely dependent upon two factors. 

 The first, and perhaps the chief of 

 these, is the color of the environment 

 into which the fishes move. The second 

 is their activity, or lack of it, for, pro- 

 vided they ever display them, banded 

 patterns appear to characterize the rest- 

 ing condition of fishes. 



It is of much interest that among 

 creatures which justly have been called 

 ''chameleons of the sea," changes in 

 color and shade should enable those dis- 

 playing the changes to adjust them- 

 selves to their normal surroundings, 

 and to assume the dominant colors of 

 their changing environment, as they 

 pass from place to place. It accords 

 completely with the discovery that the 

 colors of those fishes whose pigmenta- 

 tion is essentially stable are correlated 

 with their habits and in general repeat 

 those characterizing the stations they 

 frequent. Since the conspicuousness of 

 these creatures has not been regarded 

 as of an inferior sort, their changeable 

 coloration provides a physiological test 

 of the truth of the Neo-Darwinian hy- 

 potheses advanced to explain animal 

 coloration in terms of natural selection. 

 It may be added that if the observations 

 mentioned are correct and representa- 

 tive, it seems impossible that any ex- 

 planation of the biological significance 

 of the types of coloration in question 

 which assumes that they possess more 

 than minimal conspicuousness can be 

 correct, for what system of coloration 

 could render an animal more incon- 

 spicuous than one based upon the prin- 

 ciples indicated, if to these be added 

 the countershading the fishes almost 

 universally display? 



