ON THE BRITISH EDRIOPHTHALMA. 31 



The two first articulations, without being actually fused with the anterior 

 integumentary tissues, are sometimes so closely incorporated with them, as to 

 be lost, except to close analytical observation. This is the case in the family 

 of Orchestidcc, which has long been described by authors as having but three 

 articulations to the peduncle of this antenna ; but the other two may be seen 

 to exist in the upright anterior walls of the head, of which they form the 

 largest portion (vide Plate Xll.fig. 1 = H first articulation =P second = G third 

 and fourth). A similar conclusion is almost arrived at by Mr. Dana (Part II. 

 p. SiS). He says, "C [answering to P in our figure], an area adjoining 

 the antennae, having a membranous covering and properly a part of the 

 base of the outer antennae; d [answering to H in our figure], a shelly 

 area either side of e [C], or epistome*." This shelly area he has failed 

 to perceive, equally with P, is part of the base of the outer or second pair of 

 antennae. These articulations are so closely impacted with the head as not 

 to be observable to a lateral examination of the animal, being as they are 

 absorbed into the cephalic region. It is this peculiar arrangement ol' organs 

 in this family that pushes, as it were, the whole of the anterior organs to the 

 top of the head, placing as it does a more than usual distance between them 

 and the oral appendages. 



The filamentary termination of this antenna in the Amphiphoda is inva- 

 riably solitary and generally multi-articulate. It obtains its most filamentary 

 character in the true Gammari, but in some genera the whole of the numerous 

 articulations of which it is constructed become consolidated. 



The first approximation toward the strengthening character of this organ, 

 exists in the true Amphitoe, whence, by its near allies through Podocerus, it 

 arrives at its fulminating point in Corophium and Chelura, where they are 

 completely fused into a single articulation (vide Plate XIII.). In such cases 

 they are powerful assistants in enabling the animal to climb over uneven sur- 

 faces, and probably assist in the construction of their abodes, whether bur- 

 rowing, as Chelura and Corophium, or forming tubes, as Siphonocetus and 

 Cerapus, and probably also Erichthoneus, or in building nests, as Amphitoe 

 and Podocerus ; and to adapt them more completely to their work, they are 

 often supplied with hooks towards the extremity (Plate XIII. fig. 6 a). These 

 are formed by the consolidation of some of the capillary armature into 

 strong curved spines ; the best examples that we have observed are in Podo- 

 cerus, where they must become an additional means to the power of the 

 antenna. 



In all Crustacea this pair of antennae appears persistent and generally well 

 developed ; we are not aware that there exists in any of the Gammarina of 

 this order, or among the aberrant family of the CaprellidcB, a solitary instance 

 of its being reduced to a rudimentary or obsolete form. 



This remark appears to be true of Isopoda as well as Amphipoda, if we 

 remove from each the parasitic forms, such as the Hyperia among the 

 latter, &nd Bopyrus and its allies among the Isopoda; a circumstance, which 

 induces us to believe that the second antenna is the seat of a sense which 

 undergoes but slight modifications to enable it to be equally efficient whether 

 in air or water, since the Orchestidce live entirely out of the water, as like- 

 wise several species of Isopoda. 



The mandibles, — These are the next succeeding appendages, but are 

 separated from the last by the epistome and labium. 



The former (epistome) is generally placed in the Amphipoda, vertically in 

 the anterior wall of the head ; occasionally it is produced into a spear-like 



' * The Plates to Mr. Dana's work having been published since this has been in the press, 

 •mt have only known the references to them by the text of his work. 



