BEHAVIOR UNDER NATURAL CONDITIONS 179 



the behavior of Stentor were not regulatory, becoming more fitted to the 

 existing conditions, a comparison with the behavior of higher animals 

 in learning would be out of place. But since the changes clearly are 

 regulatory, in the one case as in the other, it would be equally out of 

 place to deny their similarity, in this respect at least. 



In another important feature the behavior of Stentor falls, so far as 

 our present evidence goes, far below the level of that found in the learn- 

 ing of higher animals. The modification in the behavior induced by 

 experience seems to last but a very short time. Immediately after 

 reacting in one way, which proves ineffective, it reacts in another. But 

 a short time after it apparently reacts in the same way as at first. 1 As 

 a rule, it is evidently to the interest of an organism living under such sim- 

 ple conditions as Stentor to return to the first method of reaction when 

 again stimulated after a period of quiet, for as a rule this first method 

 is effective, and it would be most unfortunate for the Stentor to proceed 

 to the extremity of abandoning its tube without a trial of simpler 

 reactions. But the difference between behavior which is modified 

 only for a few moments after an experience, and that which is per- 

 manently modified, is undoubtedly important. The latter would never- 

 theless be developed from the former by a mere quantitative change, 

 so that the variation in duration does not constitute a difference in essen- 

 tial nature. 



We may sum up the results of the present section as follows : The 

 same individual does not always behave in the same way under the same 

 external conditions, but the behavior depends upon the physiological 

 condition of the animal. The reaction to any given stimulus is modified 

 by the past experience of the animal, and the modifications are regula- 

 tory, not haphazard, in character. The phenomena are thus similar 

 to those shown in the " learning" of higher organisms, save that the modi- 

 fications depend upon less complex relations and last a shorter time. 



2. THE BEHAVIOR OF INFUSORIA UNDER NATURAL CONDITIONS 



We have thus far dealt chiefly with the behavior of infusoria under 

 experimental conditions. In experiments the conditions are usually 



1 This matter cannot be considered definitely settled. It is exceedingly difficult 

 in practice to devise and carry out experiments which shall actually determine the 

 length of time that the modified behavior lasts. A thorough, definitely planned investi- 

 gation should be directed precisely upon this point. Hodge and Aikins (1895) report 

 that Vorticclla, which at first took yeast as food, later rejected the yeast, and that for 

 " several hours " it refused to take the yeast again. But unfortunately no further 

 details are given. We do not know whether the Vorticella was injured and took 

 no food at all, or what other conditions were present, so that we can build little upon 

 this observation. 



