INTESTINAL RESPIRATION IN ANNELIDS. 763 



much more rapid in their rate of progress. The passage of the wave in the dorsal 

 vessel had no effect on the intestine, nor did the antiperistalsis of the intestine in any 

 way aff"ect the lumen or the contractions of the dorsal vessel. Further, antiperistaltic 

 movements are frequently altogether absent, while the vascular contractions are always 

 to be observed. 



Branchiura sowerhiji Bedd. 



The circulatory system in this species has been described by Beddard (3a), and 

 also by me (51). The following are the chief points in its anatomy : — 



The dorsal vessel is here also ventrally or ventrolaterally situated in the greater 

 part of its course ; it does not lie immediately on the intestinal wall, but has neverthe- 

 less a closer relation to the gut and to the chloragogen cells than has the ventral vessel. 

 The supraintestinal vessel reaches the sixth segment anteriorly, and posteriorly ends 

 at a point not far behind segment xii. ; it is throughout in close relation to the gut and 

 its chloragogen cells. There is a rich intestinal plexus in the gut-wall, which in seg- 

 ments s.-xiii. becomes almost a sinus ; besides having connections with dorsal and 

 ventral vessels, the plexus is joined to the supraintestinal by a series of wide 

 communications. 



The hearts are two pairs, in ix. and x. ; the first pair originate above from the supra- 

 intestinal, and unite below to form the ventral vessel ; the second pair arise above from 

 the dorsal vessel, and enter the ventral vessel below. In segments ii.-viii. there is a 

 series of non-contractile lateral loops, which give branches to the body-wall ; the loops 

 in viii. originate above from the supraintestinal, in the other segments from the dorsal 

 vessel. Behind the hearts, and in front of the gills, the lateral loops are two pairs per 

 segment ; in the gill region the distribution is complicated. The parietal vessels, which 

 originate from the loops, form a plexus in the body-wall ; but no capillaries enter into 

 the surface epithelium. 



The ventral vessel of the anterior part of the body disappears on the ventral sur- 

 face of the alimentary canal about segment ix. (cf. Limnodrdus) ; the main ventral 

 vessel of the body is formed anteriorly by the union of the hearts, and is continued 

 back to the posterior end of tlic animal. 



Relation of Contractions of Dorsal Vessel to those of the Alimentary Canal. — The 

 two series are independent. My observations may be summed up by saying that the 

 dorsal vessel contracts regularly, rapidly, and frequently ; the antiperistaltic waves 

 sometimes intermit for a period, when present progress only slowly, and are much less 

 frequent than the contractions of the dorsal vessel. 



In considering the Naididae (exclusive of the genus Chietogaster), an advance in 

 the degree of differentiation of the vascular system, as compared with the ^^olosomatidae 

 and Enchytraeidse, could be followed through such a form as Nais to Branch iodrdus, 

 the end term of the series. This advance is manifested in a greater degree of 



