HABIT AND HABITAT IN THE GALATHEIDEA. 89 



This absence of high specialization along this particular line should be 

 correlated with the habitat of the species, which visits the muddiest part 

 of the shore (just below low-tide mark) only for a very short time in 

 spring. The allied species, Galathea strigosa, shows increased com- 

 plexity of structure of the hairs which guard the inhalent aperture and 

 fringe the branchiostegite. We connect the advance in specialization 

 with the fact, noted by M. Bohn (oj). cit.), that " in G. strigosa, consider- 

 able movements of the carapace supplement the action of ihe scaphog- 

 nathite." This carapace-flapping would doubtless expose any epipodites 

 present on appendages IX to XI to the danger of being bruised and 

 torn — at all events, the epipodites of these segments are wanting. This 

 reduction of sweeping mechanism within doubles the need for a guard at 

 the entrance portals, which need is satisfied by the increased complexity 

 of straining-hairs described above. Thus, in G. strigosa, feebleness of the 

 scaphogTiathite has induced flapping of the carapace — this being in its 

 turn connected with reduction in the number of epipodites and correlated 

 complexity of straining hairs (see Plate 2, B, and Plate 4, F). It is in- 

 teresting to note that in this species the epipodite of appendage VIII 

 has alone survived, and this perhaps owing to its position opposite the 

 blunt angle of the branchiostegite, where in all probability friction is not 

 great. 



Galathea intermedia, like G. strigosa, has suffered reduction in the 

 number of its epipodites (though we are as yet unable to guess at the 

 biological significance of this reduction) ; here, only epipodites VIII and 

 IX remain, and here again correlated specialization of the straining- 

 hairs is observed (Plate 4, E). 



Two deep-water species of Galatheidea — G. nexa (to 70 fathoms) 

 and Munida subrugosa (to 600 fathoms) — while resembling G. squamifera 

 in their possession of the full number of epipodites (series VIII to XI), 

 yet afford a parallel with the case of G. strigosa and G. intermedia in the 

 relatively complex structure of the straining-hairs. This apparent 

 anomaly seems to point to a need for special precaution against choking 

 of the branchial passage in these two deep-water species ; this is one 

 item of a lengthy list of peculiarities which characterize the branchial 

 apparatus of deep-sea members of widely dissociated groups (compare 

 the peculiarities of the branchial organs of the Lepetidse among Proso- 

 branch Gasteropods, etc. etc.). 



A striking feature of such shore-living types as Porcellana is the 

 complete loss of the epipodites of appendages VIII to XI. This reduc- 

 tion should perhaps be correlated with the marked development of the 



