INTESTINAL RESPIRATION IN ANNELIDS. 749 



forwards as far as the base of the cerebral ganglion, where it bifurcates. The branches 

 are continued downwards and backwards on the wall of the buccal funnel to unite in 

 the ventral vessel, which is intimately connected with the alimentary canal for the 

 greater part of its extent, appearing on the intestine as a median ventral sinus of the 

 intestinal wall, and at the posterior end of the animal losing itself in the vacuolar 

 spaces of the gut-wall. 



Relation of Contractions of Vascular System to those of the Alimentary Canal. — 

 The vacuole-like spaces in the intestinal wall are occluded by the passage of the 

 antiperistaltic waves, which, as they pass along, constrict at the same time the lumen 

 of the gut. The vascular network, however, may at times be seen to be independently 

 contractile in a postero-anterior direction ; that is, the diameter of the lumen of the gut 

 may be quite unaffected, so that it is impossible to speak of an antiperistalsis of the gut- 

 wall, each contraction, as it passes along, merely occluding the lacunae. In still other 

 cases the movements might be interpreted either as contractions of the vascular net- 

 work, or as a feeble antiperistalsis of the gut-wall — since the waves, in addition to 

 emptying the network as they pass, cause also a slight constriction of the lumen 

 of the intestine ; and the various grades of the phenomenon seem to shade into 

 each other. 



There is often a definite relation in time between the antiperistalsis of the stomach 

 and the postero-anterior contractions which pass along the "heart," as the dorsal 

 vessel^' on the oesophagus may be named ; in the sense that these latter seem to be 

 a continuation forwards of the former — that the antiperistaltic wave, arrived at the 

 anterior end of the stomach, occasions the contraction of the "heart," which thus 

 follows it up and seems to be its extension forwards. But this relation is by no means 

 constant. Thus — 



(1) The contraction of the " heart" may not follow on every antiperistaltic wave, 

 but may occur after about every second wave. 



(2) Or, after following regularly on the antiperistaltic wave for some time, it may 

 then tend to anticipate the arrival of the latter. 



(3) Or, while still occurring definitely in a postero-anterior direction, the contrac- 

 tions of the "heart" may be quite independent of the antiperistaltic waves in the 

 stomach both in time and rhythm. They may be far more numerous, e.g. about five 

 contractions of the " heart" for every antiperistaltic wave in the stomach wall. 



(4) Or again, the direction of the contractile wave through the " heart " may in 

 some cases be definitely antero-posterior instead of, as usually, the reverse. 



In JEolosoma hemprichi, then, a condition exists in which the circulatory system 

 is in close relation with the alimentary, appearing in far the greater part of its extent 

 as a component part of the alimentary wall. Even on the oesophagus the dorsal 

 vessel retains traces of its origin as a specialised channel in a system of vacuole-like 

 spaces, since it shows us a lumen traversed by strands or incomplete septa. 



* Of. Vejdovsky (54, p. 21), " The dorsal vessel appears as a large, strongly pulsating heart." 



