(657). The blood, after being subjected to the influence of oxygen in the branchial appendages, is returned by other transverse vessels which run along the interannular septa to the alimentary canal, where they ultimately discharge themselves into the large median trunks (I) situated upon the dorsal aspect of the intestine.

(658). Considered generally*, it will be perceived that the distribution of the vascular trunks in the Eunices is pretty much the same as that which exists in the Terebellae; but when their functions are considered, and the relations in which they stand relatively to the respiratory apparatus, very important differences are at once apparent between the two genera.

(659). In Eunice, it will be seen (fig. 123) that Professor Milne-Edwards has described and figured the branchial vessels as ampullated soon after the origin of each from the common trunk, the ampullae being designed to fulfil the function of branchial hearts. These vessels therefore, according to the representations of Milne-Edwards, are, in Eunice, the analogues of those remarkable cardiac vessels (pulmonary hearts) described by Duges in the Leech. The existence of these latter vessels has already been found to be extremely problematical (§ 583.) According to the observations of Dr. Williams, these vessels in Eunice present nothing approaching to the ampullae figured in the illustrations of Milne-Edwards. The pouched dilatations are produced by the dissection and exposure to atmospheric stimulus, just as in the Earthworm the moniliform character of the descending vessel was shown to be caused by the stretching. In Eunice the lateral large segmental branches are relatively large at first, but soon divide into three lesser branches, of which one goes to the feet, the other to the intestine, and the third to the branchiae, from which the blood returns into the dorsal vessel, which in this worm, accordingly, carries arterial blood.

The suddenness of this division favours the imprisonment of a drop of blood in the first stage of the vessels, the drop thus enclosed occasioning a bulged enlargement; but this appearance is altogether accidental. The blood is admitted into and returned from the branchiae by alternate movements of contraction and dilatation: these movements are not simultaneous in all the branchiae, but various and independent in each individually, the afflux into one being synchronous with the efflux of blood from those contiguous. This contractile power is by no means peculiar to these vessels. The motion of the blood in the vessels of every part of the body of the Annelid is effected, not through the agency of uniformly-travelling undulatory contractions of their coats, but by complete contractions and relaxations of successive portions of the tube; so that, during the instant of contraction, the cylinder of the vessel in the part contracting is completely emptied of blood, the sides collapsing and meeting in the axis; and during the period of dilatation the same portion of the vessel becomes densely distended with blood: and this is the true mechanism of the circulation in those species, even, in which a central propulsive organ exists - for example, in Nats and Arenicola. In no part of the system, therefore, is the superadded contractile bulb required as an agent of circulation, since this contractile power resides in every part of every vessel in virtue of the muscularity of its parietes.

* The following axioms may be laid down relative to the circulation of the blood:

1st. In all Annelids, the blood flows in the great dorsal trunk from the tail towards the head.

2nd. In all Annelids, the blood flows in the great ventral trunk from the head towards the tail.

3rd. In the whole integumentary system of vessels, the blood moves from the great ventral towards the great dorsal trunk: this movement constitutes the annular, or transverse circulation. The main current of the blood in the ventral trunk pursues a longitudinal course until exhausted by successive lateral deviations.

4th. In Annelids, the intestinal system of vessels consists of four longitudinal trunks: one dorsal, winch may be called dorso-intestinal; one ventral, which may be distinguished as the sub-intestinal; and two lateral. These several trunks are joined together by circularly-disposed branches bearing a dense glandular capillary system. In the inferior intestinal system the general movement of the blood is from before backwards; in the circular branches, from the ventral towards the dorsal trunk.

(660). A general survey of the circulation in Eunice will suffice to satisfy the physiologist that no part of the system contains pure arterial, and no part pure venous blood. Into the double dorsal trunk arterial blood is poured from the branchiae; but to the same trunk the intestinal branches contribute venous blood: the mingling of these two classes of currents in the same trunk must result in blood of an intermediate quality. It is, then, manifest that the great subneural trunk, which in this worm is both systemic and branchial, must distribute blood of composition intermediate between venous and arterial. No part of the circulatory apparatus therefore contains pure arterial blood except the efferent branches of the branchiae.

(661). In some Annelidans we still find that gemmation performs a very important part in the reproductive process: the multiplication of the individual segments of the body depends entirely upon this mode of increase. But this is not all: it not unfrequently happens that when these animals have attained their full growth, a constriction becomes apparent near the posterior extremity, immediately behind which a proboscis and eyes are developed, forming the head of a new animal subsequently to become separated by spontaneous fissure (fig. 124); and even as many as six of these strangely-formed offsets have been counted by Milne-Edwards in continuity with each other*. The process of division is represented in fig. 124: the hinder part of the body, including about seventeen segments, is seen to be gradually separated from the anterior or larger portion; and moreover, at the point of separation, a new head, with eyes and tentacular cirri, is distinctly formed. "In one case," says Muller, "I found a mother to which three foetuses of different ages adhered in one length. The mother had thirty pedate segments: the youngest daughter, or that nearest the mother, had eleven; but the head was not yet developed. The most remote had seventeen rings, with both head and eyes, and, moreover, the tail of the mother; the middle one had seventeen segments and a head.