64 W. C. ALLEE. 



structure it would seem entirely probable that such long bodied 

 animals as Hydropsyche larvae, dragon-fly or may -fly nymphs 

 and stone-fly nymphs would show this correlation to some extent. 

 Among fishes, the bottom resting darting fishes of the type re- 

 presented by Boleosoma and Etheostoma should also show this 

 correlation. Further experimentation on these forms would 

 probably demonstrate, as the experiments here reported for 

 isopods, that the sign of the rheotactic reaction continues to be 

 of prime importance in the problem of maintenance of position 

 in streams long after the rheotactic movements have ceased to 

 be significant. 



How much of this behavior of stream isopods is an adaptation 

 to life in streams? In order to get the problem clearly stated 

 perhaps a recapitulation of relationships is permissible. By means 

 of their highly positive rheotactic reactions the stream isopods 

 are better enabled to maintain themselves in the stream environ- 

 ment through the interaction of the rheotactic and thigmotactic 

 reactions. This tendency to give positive rheotactic reactions 

 is much weaker in pond isopods and can be modified within certain 

 bounds in both pond and stream mores by varying the oxygen 

 tension of the water. Therefore it seems that this is not a 

 specific adaptation for stream habitation which the isopods 

 exhibit, but rather that adaptation (perhaps better called prop- 

 erty), of all living matter namely, irritability (Mathews, '13). 

 Beyond that, the isopod is no more adapted to the stream en- 

 vironment by reason of its positive rheotaxis than the stream is 

 adapted to the isopod by virtue of presenting the complex of 

 environmental conditions (high oxygen and low carbon dioxide 

 tension) best calculated to call forth and maintain positive 

 rheotaxis. Environmental conditions automatically cause an 

 isopod to give those responses which fit it to maintain its position 



in the stream. 



SUMMARY. 



1. The distribution of the isopod Asellus communis cannot be 

 accounted for by its rheotactic reaction alone, but can be ac- 

 counted for by the interacting thigmotactic and rheotactic 

 reactions. 



2. The movements given in rheotactic reactions of isopods are 



