The primitive alimentary tract so formed becomes divided into oesophageal and gastric regions (fig. 93 B, _oe._ and _st._). Even while the invagination of the archenteron is proceeding, the larva becomes ciliated throughout, and assumes a somewhat conical form, the apex of the cone being opposite the flat ventral surface on which the mouth is situated (fig. 93, A and B). From the apex a flagellum projects in many forms, giving the larva a helmet-like appearance. In other forms a bunch of long cilia takes the place of the flagellum (fig. 94), and in others again the flagellum is not represented. After the completion of the invagination a lobe grows out on each side of the mouth, and less well-developed lobes may appear anteriorly and posteriorly. Round the edge of the ventral surface a ciliated band makes its appearance. Two pairs of invaginations of the skin, just as in the type of Desor, now make their appearance, one pair in front of and the other behind the mouth (fig. 93 B, _pr.d._ and _po.d._), and each of them by the closure of the opening of invagination forms a sack, the outer wall of which becomes very thin and the inner wall (corresponding with the whole invagination of the type of Desor) very thick. The inner walls of the four thickenings, which I may speak of as discs, now fuse together, each disc first uniting with its fellow, and finally the two pairs uniting. [FIG. 94. A. PILIDIUM WITH AN ADVANCED NEMERTINE WORM. B. RIPE EMBRYO OF THE NEMERTEA IN THE POSITION IT OCCUPIES IN PILIDIUM. (Both after Bütschli.) _oe._ oesophagus; _st._ stomach; _i._ intestine; _pr._ proboscis; _lp._ lateral pit; _an._ amnion; _n._ nervous system.] A ventral germinal plate is thus established, which gradually grows round the intestine of the Pilidium to form the skin of the future Nemertine. The outer thin layer of each of the discs grows _pari passu_ with the inner layer, and furnishes an amnion-like covering for the embryo which is forming within the Pilidium (fig. 94, _an_). In connection with the young vermiform Nemertine there is formed on each side an outgrowth from the oesophagus (fig. 94) which is eventually placed in communication with the exterior by a ciliated canal[93]. The proboscis arises as an hollow invagination at the point where the two anterior discs fuse in front. [93] This is the view of both Metschnikoff (No. 202) and Leuckart and Pagenstecher (No. 201), and is further confirmed by Barrois, but Bütschli (No. 193), though he has not observed the earliest stages of their outgrowth, believes them to be invaginations of the Nemertine skin. When the young Nemertine has become pretty well formed within the Pilidium it becomes ciliated, begins to move, and eventually frees itself and leads an independent existence, leaving its amnion in the Pilidium which continues to live for some time.