NOTES ON STARFISHES. 215 



curling the arm E tightly round the hinder part of the abdomen 

 of the fish. Fig. 2 may be taken to represent the position of 

 the Starfish between 10.15 and 11 a.m., but it was constantly 

 making slight alterations in its attitude, and at no time was it 

 quite still. 



The other Starfish, which I will call B (dotted in the figures), 

 had now begun to move towards the Pipe-fish. Starfish B had 

 to travel about two feet in order to reach the Pipe-fish, and it 

 moved so directly towards the latter as to leave little doubt 

 that it had smelt the dead fish. At about 11.30 Starfish B 

 reached, and began to press over and against, Starfish A 

 (black in the figures) in the way shown in fig. 3, in which 



the arrow indicates the direction in which B had moved towards 

 the food. Starfish A seemed to be so much alarmed, or annoyed, 

 by the persistent pushing of B, that within a minute or two of 

 the arrival of the latter the former quitted the Pipe-fish and 

 retreated to the floor of the aquarium. 



Starfish A, however, only went about three inches away from 

 the side of the aquarium, and then began to move to and fro, 

 parallel with the body of the Pipe-fish, in a way which suggested 

 that it was too much attracted by the smell of the fish to leave 

 it, and by 12.15 p.m. it had wandered back again to the food. 

 Both Starfishes must now have released their hold on the wall 

 and retained a grasp only of the Pipe-fish, for the asteroids and 

 the fish toppled over and fell to the bottom, where they assumed 

 the positions shown in fig. 4. 



Once more, though precisely at what time I do not know, 

 Starfish A moved away from the Pipe-fish, and at 1.10 it was at 



