The gastrula opening in both Lumbricus trapezoides and Criodrilus is placed on the ventral surface, and eventually narrows to form the mouth or possibly (Criodrilus) closes at the position of the mouth. In Lumbricus trapezoides the oral opening is at first lined by hypoblast, and in Criodrilus is bounded anteriorly by three large peculiar epiblast cells, which are believed by Hatschek to assist in absorbing the albuminous fluid in which the eggs are suspended. These large cells are eventually covered by the normal epiblast cells and subsequently disappear. In both these types the hypoblast cells undergo, during their invagination, peculiar changes connected with their nutritive function. In Euaxes (Kowalevsky) the segmentation is far more unequal than in the other types; a typical epibolic invagination takes place (fig. 140), and the blastopore closes completely along the ventral surface. [FIG. 140. TRANSVERSE SECTION THROUGH THE OVUM OF EUAXES DURING AN EARLY STAGE OF DEVELOPMENT. (After Kowalevsky.) _ep._ epiblast; _ms._ mesoblastic band; _hy._ hypoblast.] In all the oligochætous types, with the exception of Euaxes, where the blastopore closes completely, the blastopore becomes, or coincides with the mouth. In Serpula it is stated (Stossich), as we have seen, to coincide with the anus: a statement which receives confirmation from the similar statements of Willemoes-Suhm (No. 358). It is necessary either to suppose a mistake on the part of Stossich, or that we have in Chætopods a case like that of Gasteropods in which a slit-like blastopore originally extending along the ventral surface may in some forms become reduced to a pore at the oral, or in other forms at the anal extremity. So far only two germinal layers--the epiblast and the hypoblast--have been spoken of. Before the invagination of the hypoblast is completed the mesoblast makes its appearance in the form of two bands or streaks, extending longitudinally for the whole length of the embryo. These are usually spoken of as germinal streaks, but to avoid the ambiguity of this term they will be spoken of as mesoblastic bands.