ANATOMY OF 1'IM.HCYPODA l:ri;\i:. 



just described, and the same is probably also the ruse in Amitiini /v>//w/</, briefly 



described hv \Yoodward (21, p. '-.'''). 



At sonic ilislaiicc in front of ihc hinder limit of the vi-veral mass tin- gill-axis 

 heroines free of the l>o<ly, ami pa es back to the anterior edge of the siphonal septum. 

 to which with the rest of ihe hinder extremity of the gills it is attached. 



The margins, both of tlie relleeted lamina of the inner demibranch and ol' the 

 upturned outer demibraiich, are attached throughout their length to the hod\ (Fig. 20, 

 <'.<>.<!.. it. i.il.). This is nut the, condition usually associated vrithAnatina; in this genus 

 tlie "'ill-edges are free as a rule, except behind the foot, though in other Anatinacea. 

 as for instance Myochama and C"''/i/<"/' *nt<t (2, p. L'S'.I). and /,////./'</ and Lyonsiella 

 (10, PI. XYI1I, fig. G'J ; PI. XIX, fig. 75), tliey are attached to the body, as in this 

 particular species of Aimlimi. 



Behind the \isceral mass th" edges of the relleeted lamina- of the inner demiliraiieh 

 are united : but the edge of the upturned outer demibranch seemed, so far as the 

 condition of this reion of the liodv allowed of reliable observation, to be for a short 



O 



distance free, leaving a narrow slit-like passage (Fig. 'JO, ".), between itself and the 

 mantle, through which tlie supra- and sub-branchial chambers of the mantle-cavity were 

 in communication. Further back, the separation of these two chambers wa- again 

 completed by the union of the edge of the outer demibranch with a forward lateral 

 prolongation of the siphonal septum (Fig. -0, *..*.). 



"With reference to this passage between the two divisions of the mantle-cavity, it is 

 interesting to notice that a somewhat similar passage, is mentioned, though witli 

 hesitation, by Hancock (2, p. 290), in his account of the gills of Myochama. lie says, 

 "In Mi/i'i'li'iiiKi. however, the branchial and anal chambers are not perfectly divided, 

 as they are in Coc/tli"/t'*iini. and probably in Pholn<l<>ni>/<i. In the former, the ^eptmn, 

 which cuts off the communication, does so only partially, there being a considerable 

 aperture in it (Fig. I,/.), .i llst where it joins the extremity of the gills. It is iptite 

 pos.-.ible that this aperture may be the result of injury," etc. 



In their minute structure the gills resemble in all important particulars those of 

 the Anatinacea described by Kidcwood (17, p. '2C,l). They are deeply pleated, with a 

 large number of filaments to the pleat (thirty in the inner demibranch, twenty-two in 

 the outer) ; the chief filaments at tin- re-entrant angles of the pleats have a strongly 

 concave fronral surface and. where interlaminar junctions are absent, present upon their 

 reverse face, in place of a small acute ridge, such as that figured by Kidewood ( 1 7, 

 lig. 51) for Clavagella, a large cylindrical prominence enclosing a capacious vessel. 

 A similar prominence occurs also upon all the chief filaments of the outer demibranch. 



At the apex of each pleat several of the lilaments (usually three) are considerably 

 larger than the normal, but \\ithoui any particular modification of shape. A similar 

 condition is said by Kidewood (17, p. -_'(i5) to be charad. ristic of the gills of Pandora 

 and Aiint!/ii/ ]!, ,< Ha.-;!, bill not of . I i/ntiini i/-nii,-,ii<i. In the gill of Chamostrea & a 

 enlarged lilamenl in this position is ligiin d by Hancock (3, PI. IN , lig. 5). 



