KEACTIOX IN ASTKlllNA GIBBOSA. 431 



From these experiments it seems clear that the primary aim 

 of the righting reaction is to bring the tube-feet into contact 

 with a solid and j'esistant surface. The first thing Asterina does 

 when detached from its hold is to stretch out its tube-feet and 

 feel roiuid in all directions. Mere tactile contact is, however, not 

 sufficient, as Experiment III. shows — the surface must resist a 

 pull. The perception involved in the contact reaction is not so 

 much tactile as kintesthetic. It does not seem to matter very 

 much whether the siu-face of adhesion lies below the starfish, as 

 in the normal righting reaction, or above it, as in Experiment lY. 

 Thei'e is no pronounced tendency to take up a definite position 

 relative to gravity. Under natural conditions Asterina may be 

 found either adhering back downwards on the underside of a 

 stone or back upwards on the bottom underneath the stone, or 

 again wedged along the lateral edges of the stone. Experiment I., 

 however, indicates that Asterina prefers to turn its ventral surface 

 to the bottom rather than crawl on a small inverted stone, and 

 this fact might be adduced, together witli the similar results 

 obtained by Preyer [ibid. p. 116), in favour of graviperception 

 taking a part in the reaction. On the other hand, Asterina 

 may crawl off tlie piece of slate simply because of its small area. 

 (The result of Experiment II. is best explained as due to that 

 " persistence of an established impulse," which is a characteristic 

 feature of the behaviour of many lower animals (Jennings, ibid. 

 pp. 115-6, 145-6). 



The probability is that graviperception in the starfish, if really 

 existent, is purely kinsesthetic. It is well known that when a 

 starfish is Avalking undistui-bed over a horizontal surface the 

 tube-feet do not attach firmly to the bottom but act A'ery much 

 like legs which push the animal forward. This has been clearly 

 shown by Jennings for Asterina forreri, and is true also of 

 Asterina. If, however, Asterina is irritated it huddles down and 

 grips the bottom firmly with its suckers. Also when at rest it 

 holds on fairly tightly. Similar]}^ if it is mounting a steep slope, 

 or adhei'ing to the under surface of a rock, it must hold on tighter, 

 bringing its suckers into play. Graviperception in Asterina. if 

 one can properly use the term, must be little more than a dim 

 perception of the difference between a surface on which it may 

 walk and a. surface to which it must cling. This gives the 

 possibility of a choice between a back downwards and a back 

 upwards position. Space to the starfish must be a tactile or 

 better a kina^sthetic space. 



To sum up, the two factors provisionally distinguished above — 

 contact and gravity — probably resolve themselves into one, for 

 both appear to depend upon kinfesthetic impressions. The aim 

 of the reaction is to geb a firm foothold on something solid ; once 

 this is found the starfish is probably able to distinguish by 

 muscle sense, in a rough and ready way, whether the surface is 

 horizontal, vertical, or inverted. 



It is possible that a third factor plays a part in the orientation 



