204 ISAO IIJIMA. 



and the inner cellular stratum of the ovarial tunic, I cannot 

 deny that there is a possibility that the cells of the latter 

 enter the germinal mass ; but I can say that I have never 

 been able to find any evidence whatever in favour of such a 

 view. In sections of the rachal side of the ovary-wall there 

 is always to be seen a distinct boundary line between the 

 germogen and the inner stratum of the tunic ; and this is 

 true whatever be the stage of development of the germogen. 

 Fig. 4 shows this line under even the youngest, least 

 tliickened portions of the germogen. In this figure the 

 central cell-string is already beginning to get loosened from 

 the wall, as is shown by the open nature of the attached 

 side. I am inclined to think that when once the massive 

 germogen has been formed, the enlargement of the strings 

 is mainly due to the growth and multiplication of its own 

 cells, and that the fusiform epithelial cells only take part in 

 forming the very youngest portion of the germogen. 



In regard to whether the germogen consists of cells, or 

 represents merely a nucleated mass of protoplasm (syncy- 

 tium), it must be confessed that the appearances are not all 

 on one side. In the fully formed free egg-string the cell- 

 limits towards the ends are quite distinct, as is shown on 

 the left side of fig. 11, PI. XVIII. In the central portions, 

 where egg-formation is going on, the outlines are sometimes 

 distinct, and sometimes apparently wholly obliterated. But 

 as this obliteration is limited to the egg-producing portions, 

 we may regard it as an evidence that some of the cells are 

 in process of absorption. Such cells would appear to lose 

 distinct outlines and to coalesce, thus forming a protoplasmic 

 matrix, in which are seen, besides the nuclei of the original 

 cells, germ-cells and ova in different stages of development. 

 The protoplasmic portion thus formed — seen abundantly in 

 figs. 12, 13, and 14 — serves undoubtedly as food for the 

 proper germ-cells and the definite ova. If we look at the 

 younger portions of the germogen, seen in fig. 4, we find 

 the cell outlines very ill-defined ; but this may be considered 

 the fault of the preparation, since in other cases, as in 

 fig. 11, such outlines are quite distinct. In this figure, as 

 in fig. 4, portions of the strings are seen to be very loosely 

 attached to the ovary-wall by a meshwork of protoplasm, 

 the meshes growing larger as the strings become heavier. 

 This protoplasm may also be regarded as broken-down cell- 

 substance. The geini-cells, of which the germogen is 

 largely made up, possess large oval nuclei, usually very 

 distinct in outline, in which one to several nucleoli are gene- 

 rally seen. In the central portions of the egg-strings, even 



