1907] Jennings.— Behavior of the Starfish. 139 



The single ray releases its suckers, and the animal tiirns by a 

 unified impulse toward the other side. (It is not at all the case, 

 as a rule, that the two rays by main force pull the opposed one 

 from its attachment; all work together. Such a violent pulling 

 away of a certain ray occurs in very rare cases, as will be men- 

 tioned later, but it is something quite out of the common.) 



(9) Kelativc number of tube feet that become attached in 

 the different rays. — Often some of the tube feet of all the rays, 

 or of a number of rays that oppose each other, become attached 

 before the unified impulse appears. Then, other things being 

 equal, the direction of turning is deteniiined by the relative num- 

 ber of tube feet attached. The rays having fewer tube feet 

 yield, release their suckers, rise up, and turn toward the other 

 side. Like any of the other factors, this one is often overcome 

 by others. Frequently a starfish is seen to become well attached 

 by one or two rays. Now another ray that has been waving 

 about finds the bottom ; its tube feet take hold one after another, 

 and now the other rays yield, release, and turn toward the one 

 later attached. There appears to be a predetermination of some 

 sort in favor of the action of a certain ray, if it can but find a 

 point of attachment. Indeed, as we shall see later, this is clearly 

 the case. 



(10) Coordinated position and action of adjacent rays. — The 

 direction of turning is in some cases determined by the way 

 that the tips of adjacent rays have twisted in becoming attached. 

 There are three possible cases here: (a) The two adjacent rays 

 may twist in opposite directions, so that their ventral surfaces 

 face away from each other (as in the two rays to the right in 

 fig. 18). Now the attached tube feet of the two rays of course 

 pull against each other, so that it is impossible for the starfish 

 to turn on these two rays, unless one or both untwist and take 

 a new position, (b) The two adjacent rays may twist both in 

 the same direction, so that the dorsal side of one faces the ventral 

 side of the other. Turning on these two rays is then very diffi- 

 cult, involving a sort of whirling movement of the body; but it 

 sometimes occurs, (c) The two adjacent rays may twist in such 

 a way that their ventral sides face each other (fig. 13). This is 

 the most favorable position for turning, the pulling of the at- 



