238 "TERRA NOVA" EXPEDITION. 



with one another in the ventral parts of the visceral mass. The genital duct passes 

 backwards along the lateral surface of the cerebro-visceral connective, and opens below 

 it slight Iv in front of and quite separately from the renal aperture (Fig. 1, </..). 



d'/'lls. The gills are supported on either side by a muscular axis, which extends 

 freely behind the adductor (Fig. 1. </..). They have the filamentary structure and 

 general freedom characteristic of the Arcacea. The two demibranchs are unequal; the 

 inner, which is the larger of the two, extends forward as far as the inner palp (Fig. 3), 

 a short distance in advance of the anterior extremity of the outer demibranch. 



The anterior filament of the inner demibranch is direct, swollen, and slightly 

 elbowed (Fig. 3, f. 1 ), as it is also in Adacnarat and P///lo/>n/a (see below). The rest 

 of the filaments in both demibranchs. except for a. few at the extreme hinder end, are 

 reflected, but have no inter-filamentary unions of any sort, except that possibly there 

 may be a ciliary union between them at the angles of reflexion and at the reflected free 

 extremities, but of this I could not be quite certain. The individual filaments are 

 remarkably short and stout, with knobbed rather than hooked free extremities, and in 

 these particulars present a somewhat embryonic appearance. The cavity of each is 

 divided longitudinally by a delicate septum. 



2. AJunnirt-,1 nifi'iis, Pelseueer. PI. II, figs. 8-9< c . 



Stations 194, 31G, 338, 339, 340, 356. A dozen or so specimens (in six separate 

 bottles) taken from 50-250 fathoms in and near McMurdo Sound. 



The anatomy of this species has been described by Pelseneer (12, p. 41), but 

 although in general the present specimens correspond very well with his description, 

 there are a few points of difference sufficiently remarkable to make it a matter of some 

 difficulty to reconcile them completely with his account. 



For instance, the "Belgica" specimens had two adductors, though the anterior 

 one was small; in these the anterior adductor is absent. In the "Belgica" specimens 

 the filaments of the inner demibranch were direct onlv : in these, except for a few at 

 the hinder end of the demibranch, they are reflected. In these specimens, too, there is 

 a well-developed "cephalic" eye, and though of course it is possible, it is at the same 

 time very unlikely that this organ, if present in the " Belgica " specimens, should have 

 escaped Pelseueer's notice, yet he makes no mention of it. 



These differences of structure are difficult to explain, and if no mistake of 

 observation or identification has been made, seem to suggest that two closelv allied 

 species may secrete shells of practically identical form. 



The examination of two specimens by means of serial sections, and of three others 

 by partial dissect ion, showed that in general structure Adacnarca niti'iift very closely 

 resembles Lix^in-n. with strong leanings in various particulars to Philobrya (see below). 



Muxi-li'x. \ posterior adductor only is present. The pedal retractors are similar to 

 those of I/iKxuri-n. except that the body-muscles are not developed as in that genus to 

 form a definite pedal elevator. 



