The body of the embryos of Aplidium and Amaroucium is formed (at an early stage) of a thick, homogeneous, external layer, and a yellowish mass enclosed within it. In the spherical portion of the body, this mass is ap parently wholly composed of round cells, of different sizes, and containing nuclei, and, towards the interior, probably unaltered glo bules produced by the division of the vitellus. These two elements do not compose any dis tinct organ ; but only form two layers, one internal and opaque, and the other external and diaphanous. In Amaroucium Nordmanni there is no canal within the tail, but its centre is occupied by a simple series of large rectan gular nucleated cells, producing a transversely radiated appearance, visible even when mode rately magnified ; the tail has also an external simple layer of minute cells, as in Botryllus.
Savigny, also, was led by his researches to regard both the ovum of Botryllus and that of Pyrosoma as giving birth to several individuals having already a certain order of arrangement. This view, in the case of the Botryllus, is, as we have seen, supported by Lawig, and Sars, and is considered by Van Beneden as founded on fact and supported by analogy. Milne-Edwards, however, is not disposed to admit this conclusion ; for, in his opinion, the existence of the four embryos united in a circle in the Pyrosoma, and the development of a single star of germs in the larval Botryl lus, do not sufficiently account for the associa tion of many such groups in the adult age ; there being, for instance, in the adult Pyrosoma, many hundred individuals of different de grees of development. Van Beneden thinks it probable that the presumed aggregate larvm produce colonies similar to themselves by fis siparous reproduction; but Milne-Edwards gives it as his confirmed opinion that, from the single product of the ovum, the other asso ciated individuals arise by gemmiparous repro duction only. Of this mode of generation in the Botryllidce we will now proceed to give a slight sketch, again acknowledging the labours of the learned Professor of the Garden of Plants as the chief source of our information. • In Biazona, Dideinnum, Botryllus, and Bo trylloides, the common test is traversed by numerous ramifying prolongations of the inner tunic of the individual animals, terminating either simply in culs-de-sac, or swelling out into germs (fig. 785.). Savigny figured these tubular bodies, and vaguely described them under the name of "marginal tubes " and " vascular branches." And Delle Chiaje#, in treating of Polyclinium, figures and describes " vessels " which are probably these ramifying canals. These membranous tubes, with their terminal vesicular enlargements, are readily seen during life in those species that possess a semitransparent test. Milne-Edwards ob
serves that in Botryllus and Botrylloides, (figs. 785 and 783, t, t'), each of these interior ap pendages appears at first as a little tubercle on the surface of the abdominal portion of the inner tunic of the adult animal. The tubercle then becomes elongated, forming a tube, the free extremity of which is closed, its cavity communicating with the abdominal cavity of the animal from which it springs. The blood from the abdominal cavity circulates through out this cmcal tube, with a very active double current. As the tubes lengthen, they generally become ramified, and soon present swollen or claviform extremities. The circulation con tinues' active, and before long there is visible towards the summit of each terminal swelling a minute granular mass, the colour of which approaches that of the thorax of the adult animals situated close by. A little later this organised mass begins to present the form of a little Ascidian, and soon afterwards becomes a young animal, similar to those already occu pying the common mass, of which it becomes a new inhabitant. The communication be tween the mother and the young animal be comes obliterated ; but for some time yet, all the young individuals growing from the same branch remain united by their pedicle, and it is this union, apparently, that determines their mode of grouping into " systems."* In Di demnum gelatinosum, the buds growing on these prolifcrous stolons are very different in appearance from the ova expelled by the animals ; for not only did they differ in aspect and form, but their bulk is at first twenty or thirty times less than that of the vitelline mass of the ova. In the Amaroucium prolifiTum, Milne-Edwards has frequently found on the surface of a rounded mass, formed by a colony of these animals, many little filiform twigs, simple or branched, formed by a prolongation of the common tegumentary substance, and consisting of a tube closed at the end, and enclosing, in its interior, one or more embryos in different states of develop ment. These young individuals terminated inferiorly by a peduncle, prolonged in form of a slender tube into the common mass, and springing apparently from the abdominal tunic of an adult individual. This mode of propa gation by buds, which the compound Ascidians possess in common with the Polypifera, is, as we have above described, found in the Clavel linidce ; the only important difference being, that in the latter the tegumentary envelope of the young is not so largely developed as in the Botryllidee, and does not become fused with that of the adults; whence it results that the individuals springing from the same stem remain isolated, instead of being united into a common mass.