194 EDWIN B. POWERS. 



of forty minutes to determine the modification of behavior over 

 a longer period of time. It was found that there was a similar 

 rhythm of invasions of the high concentration end with a final 

 coming to rest at the low acid concentration end. 



t 



4. A Comparison of the Reactions of the Four Species of Cray- 

 fishes Tested. 



While there were noted specific differences in the reactions of 

 the four species of crayfishes tested, all sensed the carbon dioxide 

 and acetic and hydrochloric acids. Observations show that 

 propinquus gives specific reactions and orients to a gradient of 

 these substances, while virilis orients to a less degree; immunis 

 to high concentrations, and diogenes to a still less degree to both 

 high and low concentrations. Both propinquus and virilis which 

 were tested at low temperatures were affected by carbon dioxide 

 to the extent of intoxication, and virilis was affected more or 

 less by the acetic acid. All species, so far as their avoiding 

 reactions were noted, showed modifications of behavior and with 

 propinquus there was a tendency to come to rest in the low end. 



VI. GENERAL DISCUSSION. 



In reviewing the data it is seen that in each set of experiments 

 all the types of behavior are increased in intensity with increase 

 in the concentration of acid used, and the question as to the 

 relation between the cause of the different types of behavior is 

 suggested. 



In the first place there must be a gradient before there can be 

 orientation. This is in accord with Weber's law which states: 

 'The increase of the stimulus necessary to produce an increase 

 of the sensation bears a constant ratio to the total stimulus," 1 

 i. e., there must be a definite ratio between the increased intensity 

 of the stimulus and the original stimulus before there can be a 

 sensation of an increased stimulus. If the crayfishes were re- 

 acting in accordance with this law, the rating, which is the numer- 

 ical expression of the degree of the reaction when both turnings 

 and time preference are considered, should be in proportion to 

 Weber's ratio, see Table III. By turning to the experiments 



1 James, The Principles of Psychology. 



