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



lifted, and the foot remains stiff and solid. As the body is moved 

 still further forward, the tube foot can no longer push ; it begins 

 to shrink and shorten ; and now, for the first time it lifts the sand 

 grain to which it is attached. This sand grain is held and carried 

 forward for a second or two ; then it drops ; the tube foot now ex- 

 tends forward again and repeats the whole process. 



Walking takes place in practically the same way on a level 

 solid body as on loose sand, so that it is doubtless brought about 

 in accordance with the same principles. If beneath one ray of a 

 starfish walking on sand a glass plate is laid, so that the feet of 

 this one ray walk on the solid, their action is indistinguishable 

 from that of those walking on the sand. There can be no doubt 

 therefore but that the common notion that the starfish in its loco- 

 motion pulls itself forward is a mistaken one. The suckers serve 

 a useful purpose in preventing the feet from slipping. As is. 

 of course, well known, there are many starfish that have no 

 suckers on their tube feet ; they walk in essentially the same 

 manner as is described above. 



In a starfish having suckers, it is a notable and significant 

 fact that during locomotion the animal is not firmly attached to 

 the substratum. Often a quiet starfish is so attached by its 

 suckers that it can scarcely be freed by force from the object on 

 which it is resting. But if we can in any way get it to begin 

 walking, we can by qnick action pick it up with the greatest 

 ease. This in itself seems to show that the attachment of the 

 suckers is not a particularly important point in locomotion. 



It is one of the most remarkable characteristics of animals 

 that they are able to accomplish the same ends in various ways. 

 I would not, of course, for a moment maintain that the suckerr, 

 are not essential when the starfish walks up the perpendicular 

 side of an aquarium or a stone, or when it hangs from the under 

 side of a surface. But even in these situations, while the suckers 

 are necessary for attachment, the actual locomotion seems to occur 

 rather by the action of the tube feet as levers, in the way de- 

 scribed above, than as cords for pulling the animal along. 



It is evident that to understand how the animal responds to 

 stimuli, it is necessary to have a clear idea of the nature of its 

 usual movements, otherwise one is likely to be led into bizarre 



