168 A HISTOEY OF RECENT CRUSTACEA 



the 3pizoantlius americanus fixes itself. First a single 

 polyp finds lodgment, and as its basal membrane spreads 

 over the shell, buds arise from it forming fresh polyps, and 

 gradually this same membrane absorbs the shell though 

 retaining its spiral shape. In this absorption there is a 

 great advantage to the hermit, because as it grows its in- 

 crease of bulk still finds room in the yielding polyp-mass, 

 without any necessity arising for a change of domicile. In 

 the case of Catapagurus Sharreri, A. Milne-Edwards (see 

 Plate IV.), there is sometimes a triple alliance, for over a 

 colony of Epizoanthus americanus there settles itself the 

 single polyp Adamsia sociahilis. The numerous specimens 

 of Farapagurm pilosimanus that were taken in depths 

 reaching to six or seven hundred fathoms were found only 

 in colonies of Epizoanthus paguripMlus^ but those that were 

 taken at much greater depths ' were either in a very different 

 species of Epizoanthus^ in naked gastropod shells, or in an 

 actinian closely resembling, if not identical with, Urticina 

 Gonsors, Verrill.' Dr. Carl Aurivillius has recently pub- 

 lished very interesting observations on the ' Symbiosis ' or 

 living together of Hydroids, Sponges, and Pagurids. Hy- 

 dr actinia echinata is frequently found coating the outside 

 and inside of various shells that are occupied by Hermit- 

 crabs. The mode of growth is such that the hydroid not 

 only repairs in effect the damaged mouth of a shell, but 

 also frequently extends its boundaries. This is especially 

 the case in districts where shells are few, and where the 

 growing Pagurid might be put to much inconvenience to 

 find a larger lodging. By the extension of the hydroid 

 colony, which sometimes gives a quite monstrous appear- 

 ance to the shell, the hermit is saved the trouble of making 

 any change of abode. The hydroids are saved from the 

 danger and damage they would be exposed to from the 

 rolling about of an empty shell. On the inside of it they 

 do not develop any of the nutritive polyps which might 

 incommode the hermit and also suffer injury from its 

 movements, but they line the interior with a network, to 

 the satisfactory smoothness of which the hermit itself con- 

 tributes. It might be supposed that this was effected by 



