26 The American Naturatst. [January 
animals—Annelid, Crustacean, and Hydroid—are here concerned, 
forming, as it were, a triple alliance, in which each is, moreover, 
dependent upon the shell of a fourth, a Gastropod. 
Over the immense sand-flats of “ Bird Shoal” the interdepend- 
ence of various creatures is especially well illustrated, and this 
seems in part due to the fact that the entire absence of stones 
and rocks leads to the burdening of the more securely fixed ani- _ 
mals and plants by those less able to resist the changing tides, 
waves, and moving sand. Thus we find the stout tubes of the 
Annelid, Diopatra, which project several inches from the sand, 
seized upon by Algz, Hydroids, Molluscs, Annelids, and especially 
by large Ascidians, as the only available, somewhat stable founda- 
tion to build upon. Here also occurs in great abundance the 
small hermit crab (Eupagurus longicarpus), inhabiting the empty 
shells of various small Molluscs, especially the most accessible 
one,—that of the Gasteropod, //yanassa obsoleta. Quite a large 
proportion of the shells so inhabited are covered over almost com- 
pletely by colonies of the interesting Hydroid, Hydractinia, though 
the constant moving about of the crab keeps a small circular area 
of the shell free from this growth, owing to the friction of the 
shell against the sand as it is dragged along. Undoubtedly the 
Hydractinian is benefited by this association, since it may obtain 
some of the food not used by the crab, and since it is upon a 
surface kept by the crab above the bottom, and protected from 
the constant danger of permanent burial in the sand. If, on the 
other hand, the crab, as seems probable enough, is protected by 
the presence of so well-armed a creature as the Hydroid, we would 
have here a cause of “mutualism.” With this combination of — 
Hydroid and crab is also associated a commensal,—a small An- — 
nelid of the family Spionide and genus Polydora. Of several . 
hundred shells examined, fifty per cent. were inhabited by this 
Polydora, in addition to the crab. In those without Hydroids the — 
Annelid is less abundant, but such shells are often very old or 
broken, or else but temporarily used by the crab. Occasionally 
an empty shall contains an Annelid; but here we may assume 
that a crab has recently occupied it ad only been out for a short : 
time, since the majority of available shells are either taken pos- 
