444 THOS. H. MONTGOMERY jr., 
distension of the ova upon it it is stretched out to form a very thin 
membrane. This is a direct continuation of the lining of the uteri 
(Ut Fig. 64, Pl. 40), and may be termed the germinal epithelium. 
When fully stretched out before the time of oviposition it appears on 
cross section as an exceedingly thin line, and where it is closely ap- 
posed to an ovum is indistinguishable from the membrane of the 
latter. But during or after the discharge of the ova the wall of the 
ovary contracts, the tension upon it being removed, and the whole 
organ then represents a collapsed and folded sack (Ov Fig. 63, 
Pl. 40; Fig. 77, Pl. 42) attached to the wall of the uterus (U¢). At 
this stage it can be readily determined that it is completely closed 
against the lateral body cavity, all of its ova passing into the uterus 
and none of them into that body cavity. Then its lining, the germinal 
epithelium (G. Epi Fig. 77, Pl. 42), does not appear even but is much 
folded, and these folds have in general the form of rays diverging 
outwards from the point of their attachment to the uterus. And even 
before the discharge of the ova the latter can be seen to have a cor- 
responding radiating arrangement within the ovary (Fig. 66, Pl. 41). 
These rays or lobules of the ovary are further subdivided or branched 
at their ends, so that the whole organ presents a racemose arrange- 
ment when it is empty; the same arrangement of course holds also 
during its distended condition, but then the continuity of the epi- 
thelium cannot be so well determined. When empty the germinal 
epithelium of each ovary is much thicker (Fig. 77, Pl. 42), and is 
then found to consist of an epithelium with disc-shaped nuclei, none 
of them in stages of mitosis; the epithelium then arranged more or 
less in the form of a loose and irregular network, the meshes of 
which correspond to the spaces previously occupied by the ova; the 
finer cytoplasmic strands of this network evidently correspond to 
earlier individual compartments for the ova. 
The lobular subdivisions of the each ovary probably represent 
divergent terminal branches which have grown out from a single one 
arising from the lateral wall of the uterus; this would explain the 
convergence of all the lobes to one point, and is in accord with the 
observations of VEJDOVSKY on the development of Gordius. 
The ovaries are arranged close behind each other in each lateral 
body cavity (Ov Fig. 62, Pl. 40; Fig. 67, Pl. 41; Figs. 78, 79, Pl. 42); 
they touch each other when distended with ova, otherwise small 
spaces separate them, these spaces being largest between the most 
anterior ovaries; these most anterior ovaries are also the smallest, 
