REACTIONS OF CRAYFISHES. 189 



rapid modification of behavior (Chart I., Expt. 5). Periodicity 

 is still present but is marked by shorter invasions of the high end. 

 This periodicity is better shown by individuals than by groups. 

 The graph of experiment 5, Chart I., shows that there was a 

 complete cessation of invasions of the high end after 48 minutes 

 and the crayfishes were still resting in the low end after 80 

 minutes. Numerically expressed, 70 per cent, of all individuals 

 tried showed modification. 



(b) C. virilis. Virilis oriented less definitely to the carbon 

 dioxide gradient than propinqmts. (Compare ratings, Table 

 III.) The lack of orientation is shown by the percentage of 

 turnings from the two ends and the time preference for one end 

 or the other. This staggering is possibly due to the somewhat 

 more concentrated solutions used, but may be explained as has 

 been suggested by Shelford and Alice ('13) for swift water fishes. 

 That is swift water fishes which encounter very little carbon 

 dioxide may react less definitely to it than fishes which live more 

 often in the presence of carbon dioxide. Virilis showed a ten- 

 dency to crawl out of the water. 



(c) C. diogenes. Diogenes reacted less intensely to both the 

 lower and higher concentrations of carbon dioxide in gradients 

 than either propinquns or virilis. In experiment 44 with 18.9 c.c. 

 carbon dioxide per liter at the high end there was no backing 

 accompanying the turnings, while in experiment 28, with 1 79.4 c.c. 

 carbon dioxide per liter at the high end, there were 44 per cent, 

 of the turnings accompanied by backing and one of the turnings 

 on first encounter was accompanied by backing. Diogenes also 

 showed a marked tendency in the high concentration carbon 

 dioxide gradient to move forward and then stop after which it 

 would move forward again. 



(d) C. immunis. All experiments with immunis were with 

 concentrations not running above 25 c.c. carbon dioxide per liter 

 at the high concentration end. The avoiding reactions with the 

 exception of experiment 45 were very low. There was a slight 

 positive reaction in experiment 23 as is shown in Table III. 

 There were two turnings from the high concentration end 

 accompanied by backing in experiments 23 and 46 while at the 

 same time there were 3 from the low end. 



