1^07] Jennings. — Behavior of the Starfish. 77 



podicellariae that held to small crabs for an hour or more. It 

 is very common for a small crustacean to be held immobile on the 

 back of a starfish for forty-eight hours or more, and there is no 

 indication that the pedieellariae change off or take turns in hold- 

 ing; apparently the same pedieellariae hold for the entire period. 

 Living animals seized by the pedieellariae usually try to 

 pull away, and the small pedieellariae are so constructed that 

 the pulling tends to make them hold the more strongly. The 

 two jaws are crossed like the blades of a pair of scissors, and 

 to the end of each is attached a muscular strand. These two 

 strands come together and extend straight down the stalk of 

 the pedicellaria (fig. 3). When the pedicellaria is pulled, so as 

 to stretch the stalk, these two strands necessarily tend to hold 

 the two jaws more tightly together. The stalk is extremely 

 short — hardly noticeable, indeed,^under usual conditions; it is 

 shorter than the pedicellaria itself. But it may stretch to many 

 times the length of the pedicellaria. When the object seized 

 pulls too strongly it will usually slip out of the jaws before the 

 stalk breaks. Thus the pedieellariae are rarely pulled otf, 

 though this sometimes happens. This pulling of objects away 

 from the pedicellaria is a common episode in the capture of 

 food. After a stretched pedicellaria has been released, it con- 

 tracts strongly, and remains contracted, not opening in response 

 to stimulation for a long time. 



But the long continued holding of the pedieellariae is not due 

 alone to the pulling, with its attendant stretching of the muscle, 

 as might perhaps be supposed. Minute lifeless objects are often 

 held for long periods. In one ease a single pedicellaria was ob- 

 served holding a minute lifeless spine for thirty-four minutes; 

 in another a minute appendage from a crustacean was held 

 while under observation for forty-two minutes; in another a 

 bristle from a brush was held for twenty-one minutes. How 

 much longer before and after the period of observation these 

 objects were held I do not know. IMany similar cases were ob- 

 served. As we have before noted, while thus holding an object, 

 the pedieellariae do not, as a rule, open when stimulated in a way 

 that causes the free pedieellariae to open. 



On the other hand, minute objects are often held for but a 



