70 University of California Publications in Zoology. [Vol. 4 



Some responses to chemicals are food reactions rather than 

 defensive movements. This is the case with the juice of crab 

 meat. If a little of this jnice, or a piece of the crab meat, is 

 brought near the dorsal surface of the starfish, the rosettes rise, 

 the pedicellariae open, and if possible seize the meat. The latter 

 is then carried to the mouth, in the way that will be described in 

 our account of the food reactions. 



Perhaps the surest way to cause the rosettes to attack is to let 

 another starfish brush against the specimen under observation; 

 there is usually a prompt rising of the defensive organs over a 

 large area. The touch of a single tube foot of another individual 

 often causes extensive rising of the pedicellariae; the animal 

 bristles up like a cat. Even when an individual draws one of its 

 own rays across another there is a general rising of the rosettes. 

 Contact with almost any other animal has a similar effect. In 

 these cases there is perhaps a combination of chemical and me- 

 chanical stimulation. 



In many specimens, if the animal is seized and turned on its 

 back, or is otherwise maltreated, there is a simultaneous attack 

 by all the rosettes ; they rise and the pedicellariae open. Whether 

 this occurs or not depends partly on the general physiological 

 condition of the individual. Starfish which have been weakened 

 through a long stay in the laboratory tanks commonly do not 

 react in this way, while those freshly brought in from the sea 

 are much more likely to do so. In general, I found that large, 

 old specimens are much more likely to respond by a general 

 bristling upon slight provocation than are younger ones. There 

 is much individual difference in this matter. Working for some 

 time continuously with a number of individuals, one gets an 

 impression as of the existence of mild-tempered starfish and of 

 exceedingly ill-tempered ones. A certain large starfish H, used 

 in my experiments on habit formation described later, belonged 

 to the latter class; whenever touched, its rosettes all rose 

 savagely. 



Opening, Closing and Grasping. — In the retracted rosettes 

 the pedicellariae are, under ordinary conditions, closed. If now 

 a small animal or other object comes against one of these resting 

 pedicellariae, the latter does not react at all. Before it will snap, 



