190"] Jennings. — Behavior of the Starfish. 169 



powerful and long continued action of the habit-forming agent 

 The same considerations show why a habit once formed is 

 bound to disintegrate when the starfish is left to itself. Suppose 

 that the habit tends to make the animal turn on a — b. A combi- 

 nation of other factors may induce it to turn in a given case on 

 d — e. As soon as this happens we of course get the beginning of 

 the formation of a new habit, — a tendency to turn on d — e. 

 Every lapse from the habit weakens it and tends to start a new 

 one, and the effective factors are so numerous that such lapses 

 must frequently occur. It is quite possible, indeed, that tenden- 

 cies due to other activities may form or disintegrate special habits 

 in the righting reaction. Thus, if the starfish acquires in any way 

 a tendency to creep with a certain pair of rays in advance it is 

 quite possible that this same pair of rays will tend to play the 

 active part in the righting reaction. Such possibilities deserve 

 a careful study in a systematic investigation of habit-formation 

 in the starfish. 



The many factors which may determine the righting reaction 

 will evidently have no constant tendency to make the starfish turn 

 on one given pair of rays ; on the contrary, they must sometimes 

 act in one direction, sometimes in another. It is evident, there- 

 fore, that we can hardly expect any very fixed habits to be 

 formed, under natural conditions. As we have seen, there is 

 usually some tendency in a given individual to use certain rays 

 somewhat more frequently than others ; this is as marked a habit 

 as could be anticipated. 



THE UNIFIED IMPULSE AND THE DETERMINATION OF THE VARIED 

 ACTIVITIES UNDER ITS INFLUENCE. 



We have thus far dealt with the determination of the action 

 of different parts of the body before the impulse is formed, and 

 with the determination of the direction of the impulse ; we have 

 also, in the study of habit-formation, watched the development of 

 the unified impulse. We shall now deal with some general ques- 

 tions presented by the existence of the unified impulse, and par- 

 ticularly with the way in which the action of different parts of 

 the body is determined after the impulse has appeared. 



