(662). From various observations, it would seem that similar phenomena present themselves during the development of other Annelidans; proving that the bodies of these animals grow by the successive formation of new segments, or zonules, which sprout from those already in existence, in accordance with a fundamental plan, and become arranged in regular sequence, one behind the other.

(663). It is likewise evident that the two extreme portions of the body, namely the oral and anal segments, are first formed, and that it is in the space between these that all the segments of the trunk, however numerous, have their origin, - their development being carried on in a single series, which is progressively extended from before, backwards, by the continual addition of new segments, which are so disposed that the relative age of each ring is indicated by the position which it occupies. Each newly-formed zonule is invariably interposed between the last-developed segment and the anal ring; so that it becomes natural to inquire from which of these two it more immediately derives its origin, - a question that at first might appear of difficult solution, but which seems to be set at rest by the following considerations. From the observations of M. de Quatrefages1, it appears that in some genera, at a certain period of their growth, a new individual, entirely appointed for sexual reproduction, is developed from the posterior part of the body, from which it separates itself after having remained for some time adherent thereunto.

Now these young Annelidans, thus formed by a process of gemination, are developed precisely in the same position as the new segments of the larvae, that is to say, between the anal ring and the last zonule of the trunk; but they are not all formed at the same time, and, from the different stages of development at which they had each arrived in the individual described above (§ 661), it was evident that the youngest were placed nearest to the trunk of the parent animal. The first-formed young one, therefore, had been, as usual, primarily situated between the terminal segment of the trunk of the adult Annelid and its anal segment, which, being consequently pushed backwards, had ceased to belong to the parent stock and had become a constituent part of the newly-formed young one; the second must, consequently, have been developed between that first formed and the same terminal joint of the trunk of the parent, and could have had no relation whatever with the original caudal segment, and so on for the third and fourth, etc., - proving that the penultimate segment of the body, and not the caudal ring, constitutes the point from which development emanates.

The gemmiparous production of a new individual resembles, therefore, to a certain extent, the formation of the new zonules in the body of the larva: only, in the latter case, the reproductive ring loses its creative power as soon as it has given birth to a new segment, with which it becomes intimately connected, and which, in its turn, assumes the reproductive faculty; whilst, on the contrary, in the process of multiplying individuals by gemmation, the product becomes, to a certain extent, separated from the economy of the parent animal, and the reproducing segment retains its gemmigerous faculty and gives origin to a series of new beings, one after the other, the last-formed pushing their predecessors further back as they are successively developed.

Gemmiparous reproduction of Cirrhatula.

Fig. 124. Gemmiparous reproduction of Cirrhatula.

* The phenomena of fissiparous generation in these Annelidans will be better understood by reference to Mr. Newport's important discoveries relative to the growth of the Myriapoda. - Vide post, § 732.

1 Ann. des Sci. Nat. 1844.

(664). Some curious speculations have been entertained by continental writers relative to this mode of propagation. The tail of the original Nereis is still the tail of its offspring; and however often the body may divide, still the same tail remains attached to the hinder portion, so that this part of the animal may be said to enjoy a kind of immunity from death.

(665). In Arenicola piscatorum (fig. 125), a worm met with abundantly upon our own coasts, and eagerly sought after as a bait by fishermen, who dig it from the holes that it excavates in the sand, the branchiae (6) are confined to the central portion of the body, where they form on each side a series of bunches, remarkable during the life of the creature for their beautiful red colour, derived from the crimson blood that circulates copiously through them.

(666). Respiration is performed in Arenicola by means of naked blood-vessels, projecting, in the adult worm (fig. 125, b), at the root of the setiferous processes upwards and outwards one-fourth of an inch from the surface of the body. They are limited in number and distribution to the fourteen or sixteen middle annuli or segments. These external branchiae are commonly described as forming simply arborescent tufts; the division of the vessels is, however, found, on more minute examination, to be regulated in accordance with a fixed principle. When fully injected with blood, the vessels of each branchia form a single flattened plane, which rises obliquely above and across the body immediately behind each brush of setse. In the adult animal each gill is composed of from twelve to sixteen primary branches arising from a single trunk that proceeds from the great dorsal vessel': the vessels in the branchial tuft describe zigzag outlines, - the secondary branches projecting from the salient point or the outside of each angle of the zigzags, and the tertiary from similar points on the secondary branches. This mode of division, occurring in one place and in all the smaller branches, results in a plexus of vessels of great beauty of pattern or design.