63 



groove marked A^. 4 — 6 in fïg. [52. This groove, which is of course open ventrally, may be 

 no more than a temporary told of the arm-base, but it indicates the path which the food- 

 particles- must take in passing from arms 4 — 6 along the ventral side of the arm-base to the 

 food-passage which is boundetl in front by the posterior wall of the proboscis. It thus appears 

 that mucus passing along the anterior face of the opercuknn in figs. 155 — 152 woukl be brought 

 into relation with the food-grooves of the posterior set of arms. The mucus leaving the mouth 

 by the more median oral groove in fig. 155 woukl for the most part remain on the same 

 side of the epithelial ridge e.r. (figs. 154 — 152), while the additional ridge seen between e.r. 

 and the proboscis on the right side of fig. 152 woukl keep the mucus from passing further 

 along the dorsal wall of the proboscis; and woukl, on the contrary, direct it to the region of 

 the food-grooves of the first three arms. 



The mucus which passes down the posterior wall of the pharyngeal diverticulum, whether 

 in one or two grooves, might spread itself out on the dorsal wall of the pharynx, where it 

 would be well placed to entangle food when the pharyngeal diverticulum had the position seen 



in figs. II. 37 (PI- IV). 



Some at least of the food, entangled in the mucous secretion, is probably caught by the 

 operculum and passes to the bottom of the opercular recess (op. ree.) indicated in figs. 34, 42. 

 Here I imagine it to accumulate and from time to time to be taken into the pharynx by a 

 definite swallowing action, due probably to the intrinsic muscles of the operculum (cf. Pk IX, 

 fig. T07). It then probably passes down the more median parts of the pharynx, along the deeply 

 staining and almost certainly ciliated cells which constitute both its anterior and posterior wall. 

 I think that Masterman is probably right in suggesting that the relation of the gill-slits to the 

 pleurochordal grooves is to be explained by supposing that the excess of water which passes 

 to the oesophageal end of the pharynx travels forwards along the grooves to issue by the gill- 

 slits. The relation of the grooves to the pharyngeal diverticulum further suggests that this return 

 current of water, entering the diverticulum from behind, would also have the effect of carrjang 

 the mucus forwards along the grooves of the upper lip to the point where it meets the inrush 

 of water caused by the ciliary action of the tentacles. Finally it appears to me possible, from 

 a consideration of such preparations as figs. 123 — 125 (Pk X) that the lumen of the pharynx 

 is not necessarily widely open during the whole process of feeding. It seems possible that the 

 anterior and posterior walls of the pharynx may be apposed to one another in such a way as 

 to close the cavity. This might happen during the accumulation of food in the opercular recess, 

 the water meanwhile passing directly to the exterior by the gill-slits. 



As actual evidence of the existence of mucus in connexion with the feeding process it 

 may be remarked that M'Intosh (87, p. ló) in describing the structure of the pharynx states 

 that in some preparations a film occurs on the surface of the epithelium which "is evidently 

 due to mucus". I have observed appearances in sections of C. dodecalophis which point in the 

 same direction, the film in question occurring not only along the free ends of the epithelial cells 

 of the pharyngeal diverticulum, but also along those of the cells of the anterior and posterior 

 walls of the pharynx, throughout the whole length of that organ, and even on those parts of 

 the operculum and proboscis which are near the entrance to the alimentary canak 



