On the ovogenesis and spermatogenesis of Sagitta bipunctata. 299 
Development of the ova. The youngest oocytes are always 
found developing from the median central portion of the layer of 
germinal epithelium (Fig. 1), suggesting the possibility that only that 
portion of the layer is the true germinal epithelium, and that the 
remainder, which furnishes follicle and other accessory cells, has a 
different origin, but this is a question which can be answered only 
by a study of the embryological development of the ovary. Dividing 
cells are found in all parts of this epithelial layer; in several cases 
I have found the somatic number of chromosomes, 18, in cells of 
the side wings, but have not been able to count them in the central 
region, they are always so massed together in division stages. 
As a rule, the central portion of the ovary contains only a mass 
of very young oocytes, often 10—12 in a section, with deeply-staining 
large chromosomes, a very small amount of cytoplasm, and very in- 
distinct nuclear and cell membranes. Occasionally there are found on 
the border of the spermduct cells with large granular nuclei and ir- 
regular, deeply-staining granules in the nuclear membrane, as shown 
in Fig. 3a; in the same figure are seen two ordinary epithelial cells, 
b, just outside of the ovarial region. Figures 4—8 show the youngest 
oocytes in which the chromosomes can be counted with certainty, and 
in such nine are present, very often in the form of loops of different 
length and thickness, and always with an uneven surface. These 
oocytes were all found in the central region of the ovary, some lying 
against the wall of the spermduct, others apparently detached from it. 
Fig. 4 shows a very common form, often longer, with very large black 
granules in the nuclear membrane. These granules, which are also 
seen in Figs. 6, 8, 9, are irregular in form, number, and location in 
the membrane; they become smaller and more numerous as the oocyte 
increases in size and evidently go to form the reticular network which 
is conspicuous in the nuclear membrane of all oocytes up to the time 
when the egg-membrane forms. A small portion of this reticulation 
is shown in a surface section of a nuclear membrane in Fig. 1a. 
The regular arrangement of loops in such oocytes as are shown 
in Figs. 5 and 6, indicate the possibility that they may have begun 
their development after the last oogonia division without an inter- 
vening resting stage, and that the reduction in number to nine bi- 
valent chromosomes may have recently taken place. Figs. 7 and 8 
were intended to show the expanding and lengthening loops, and no 
attempt was made to show all of the. chromosomes. 
Fig. 9 shows a section of a somewhat older oocyte in about the 
