52 



but while this is going on as we approach the ctenidial 

 axis, the same structure of the opposite demibranch is 

 drawing very near to it, until both demibranchs fuse by 

 their inner surfaces. The cavities of the ordinary 

 filaments, bounded by their chitinous skeleton, elongate 

 in a direction at right angles to the plane of the axis, 

 until they eventually come in contact with those of the 

 opposite demibranch, and thus we have the chitinous 

 skeleton of both demibranchs and the cavities of the 

 filaments continuous and all opening into the efferent 

 vessel of the gill axis. 



This elongation of the cavities of the filaments witli 

 their skeleton is probably only due to the sections passing 

 through obliquely owing to the filaments of both sides 

 curving over as the chitinous framework becomes 

 continuous. Thus the blood, which has been forced to 

 the various parts of the body with the exception of the 

 mantle, and has been collected and taken to the renal 

 organs (fig. 16, Ro.), passes from these on each side by 

 the afferent branchial vessels (fig. 43 and fig. IG, T3r. aff .), 

 and then from these laterall}^ into the vessels running 

 down the margins of the branchial expansions of the prin- 

 cipal gill filaments. From these vessels it can pass into the 

 expansions themselves, the whole of which act, therefore, 

 as respiratory surfaces. This brings the blood into the 

 principal filaments, from which it passes into the efferent 

 branchial vessel running just below the afferent. Since 

 the ordinary filaments only open into the efferent vessel, 

 the blood that passes through them must be partially 

 aerated, and there will in all probability be no definite 

 current, but a backwards and forwards motion. It seems 

 certain, from the development of vessels in the mantle, 

 that the great function of the gills is to produce currents 

 of water for aeration, nutrition, and the carrving awav 



