534 W. H. LONGLEY 



A study of the interrelations of the gulfweed fauna was in 

 progress, when the prevalence of Sargassum colors upon animals 

 living in it focussed attention almost at once upon issues raised 

 by this agreement. It appears that no less than 12 species 

 (4 fishes, 2 crabs, 2 shrimps, 3 gastropods, and a planarian) are 

 marked with the brown and yellow of the algae, wdth or without 

 the white of their adherent bryozoan skeletons and worm tubes. 

 No animals normally restricted to the floating weed distinctly 

 display other colors, and the convergent evolution of the group 

 is indicated in addition by the fact that several have organs so 

 shapen that their resemblance to parts of the plants can scarcely 

 be ignored. 



These observations suggest many problems whose direct pre- 

 sentation reinvests them with the freshness and interest they 

 possessed when pioneers of Darwinism first attempted their 

 solution. Among them few seem more fundamental, than why- 

 in some habitats all the animals should wear dull colors, while 

 in others conspicuousness should appear at a premium, and no 

 hues be too gaudy to serve in attaining it. 



It is noticeable that in the former case the animals' colors 

 tend to repeat those of their surroundings. In the present in- 

 stance one is impressed besides by the fact that agreement in 

 color is correlated with coincidence in range, and exaggerated 

 resemblance to surrounding objects with intimate association 

 with them; for the gulfweed animals are essentially inhabitants 

 of small oceanic islands, and are cut off from racial points of 

 origin by open water. But, if a tendency toward uniformity 

 in coloration characterizes groups of sluggish animals living in 

 surroundings whose color-characters are relatively invariable, 

 it is possible that the colors of more active ones living under 

 less uniform conditions may obey the same laws of distribution 

 without the fact being obvious. For in one case all the colors 

 which the animals wear may surround them at all times, while 

 in the other this may not be so, and the fact of repetition may 

 thereby escape detection. Hence, it was accepted as a working 

 hypothesis that, in general, the colors which niay appear upon 

 animals are rather strictly conditioned by the environment in 

 which they live. 



