116 "TERRA NOVA' KXPHDITFON. 



II. SYSTEMATIC AND MORPHOLOGICAL. 



Order METANEMERTINI. 

 Fam. AMPHIPURID^E. 



A/iip/il/iii/'iis. Elireiil)L'rg, 1831. 



1. ^iii/p/ii/KH'iis moselci/i, Hulirecht. 



Ampliiporus moseleyl, Hubrecht, 1887, pp. 20-22 ; PL I, figs. 20, 21 ; PI. IX, figs. 4, 7-9 ; 

 PI. X, fig. 3 : PI. XV, figs. 11, 12, 20. 



Statiuiis 220, 2U4, .3:39 : 45-158 fathoms. 



The c-ollcetion contauis three small specimeurf and various fragments, inchiding 

 pieces of the proliosi-is, whieh I assign to this speeies, not, however, without some 

 hesitatioD. The material being so scanty, and the specimens not attaining to the 

 large size of the types in the " Challenger " collection, though some of them are 

 sexually mature, it may be questioned whether we are not here dealing with a new 

 form. But on comparing the details of the proboscis-armature, and hand-sections 

 taken through the whole animal, with the "Challenger'" preparations and figures 

 of this species, the resemblances are so great, and the differences so slight, that I 

 do not feel justified in making a specific distinction. 



With so small a num])er of specimens at my disposal. I was unable to cut 

 a series of sections, and have therefore no important contribution to make to the 

 anatomy of the species. It ma}' be remarked, liowever, that in the various trans- 

 verse sections taken liy hand, chiefly in the middle or posterior region of the 

 body, the lateral nerve-stems do not appear to lie situated quite so far dorsallv 

 as in the tyjie material. IIu1)reclit, indeed, la\'s 2''^i'ticular emphasis on the point 

 that the uerve-stems lie dhm-c the lateral gut-caeca in ^1. museh'j/i . and notes this 

 as one of the characteristic features of the species. In the examples now under 

 consideration the nerves do. apparent!}-, lie at some distance from the lateral maigin 

 of the ])0(ly, and nearer to the dorsal than the ventral side, l)ut thc\- are not 

 entire!}- dorsal to the gut-cseca, parts of which extend outwards liotli al)ove and 

 licjow tliem, and may l)e said t(.) envelope tliem, as it were, on three sides. T!iis 

 sliglit discrepancy may, of course, lie due merely to dilferent states of contraction, 

 or to a difference in the level at which the sections were cut. 



The nerve-layer of the pro])oscis, as I find is the case in the ■" Challenger " 

 material, contains fourteen longitudinal nerves. 



The genital organs are also arranged as stated in Hubrecht's account. 



