222 THE INFLUENCE OF INANIMATE SURROUNDINGS. 



pouch rests when the creature has completely withdrawn itself 

 into it, displays the radial septa of a polyp-cup one above 

 another. They there are perfectly distinct, while the side 

 walls of the cylindrical cavity are so completely lined with a 

 thin calcareous crust that nothing can be seen of the perpen- 

 dicular septa of the polyp-cup. From this it is evident that 

 the young crab, or the larva of it, takes up its abode in the 

 centre of a cup, and so kills the polyp inhabiting it. A speci- 

 men now lying before me, with an incomplete cave-dwelling, 

 shows that the crab grows at first at the same rate as the 



FIG. 68. Gcniastrasa Bournoni, M. Edw., with a funnel, at the bottom of which a crab 

 (Cryptochirus comlliodyles, Hiller) is sitting, only the head being visible. 



surrounding polyps : for the margin of the crab's hole, which 

 is perfectly cylindrical, is on exactly the same level as the 

 neighbouring cups, and its breadth too is exactly the same. 

 The cavity is six millimetres long, and the length of the crab 

 found in it exactly corresponds. In another example, however, 

 the length of the pit is twenty millimetres, while that of the 

 crab belonging to it is not more than seven millimetres, at any 

 rate in the dried state. This proves that the crab ceases to 

 grow much sooner than the coral ; and this conclusion is strik- 

 ingly confirmed by the fact that the margin of the cylindrical 

 pit is not on the same level as that of the surrounding polyp- 



