1907] Jennings. — Behavior of the Starfish. 135 



direction, some in another, some remain quiet for a time, then 

 later begin to act, etc., etc. 1 



The first two questions are closely interconnected, and will 

 be discussed mainly together. The third question is more inde- 

 pendent, and will be taken up separately. 



1. As to the first question, two general facts may be pointed 

 out: (1) The action of the different rays before the impulse 

 is set is sometimes determined by internal conditions, since they 

 often show well-defined tendencies at the moment the animal 

 is turned over. This has been set forth above (p. 132). (2) The 

 action of the different rays is partly determined by external 

 agents of various kinds. Both the internal and external factors 

 are largely the same as those which determine the direction of the 

 impulse, so that it will hardly be worth while to deal with them 

 separately under each of these two heads. All the factors will 

 then be considered under the next question. 



2. Direction of the Impulse. — When the starfish is placed on 

 its back, how is the direction of turning determined? We will 

 take up separately the various factors involved. 



(1) Light.— The direction of the rays of light plays a large 

 part in determining the impulse. If the starfish is placed on its 

 back near a window or other source of light, there is a marked 

 tendency to turn, in righting, away from the lighted side. This 

 factor is, however, only one of many, so that the results of ex- 

 periments with light are of course not invariable. 



Three specimens were placed before a brightly lighted win- 

 dow, and were each inverted on a plane level glass surface, ten 

 times in succession. The animals were simply taken up and 

 dropped on their backs, care being taken that there should be no 

 constancy in the relation of given rays to the direction of the 

 light. In the resulting thirty experiments, the starfish turned 

 twenty times almost directly away from the window ; four times 

 obliquely away from the window, three times at right angles to 

 the direction from which the light came, and three times directly 

 toward the window. 



In another series five specimens were used and each was 

 given six trials, making again thirty experiments. In this series 

 there were nineteen turns away from the window, three obliquely 



