OVARY AND OVARIAN EGG OF ANOPHELES 437 
applied to the nurse-cells and gives the appearance of ingesting 
them. 
The nurse-cells, which consist merely of degenerating nuclei 
invested by the cell membrane, now pass into the epithelium, 
in which they lie till they become completely degenerated. 
It will be seen from fig. 39 that they have every appearance 
of being ingested by an epithelial cell, i.e. a rosette-cell, though 
I have been unable to demonstrate that they are completely 
surrounded by the cytoplasm of the rosette-cells. ‘This is not 
surprising as, owing to the large size of the nurse-cell and the 
comparatively small size of the rosette-cell, the layer of 
cytoplasm of the latter surrounding the former would of 
necessity be exceedingly thin, and would be very difficult 
to distinguish from the nurse-cell membrane or from the 
surrounding epithelial cells. 
Whether the degenerating nurse-cells are completely ingested 
by the rosette-cells or not, it is certain that there is a very 
intimate relation between the two, and the latter invest a 
considerable portion at least of the former. The degenerating 
nurse-cell nuclei would appear to form a general food reserve 
which is used by the rosette-cells while forming the micropyle 
apparatus. 
The large size of the nurse-cells causes the radial arrangement 
of the rosette-cells to appear distorted, though indications of 
this arrangement can always be made out. 
The micropyle apparatus arises under the rosette-cells at 
the same time that the rest of the chorion appears. The whole, 
with the possible exception of the narrow portion of the 
funnel, is secreted by the rosette-cells, and there is no obvious 
mechanism to account for the secretion of the thick supporting 
ring by part of the surface and the thin disk by another. 
The bases of the epithelial cells which are surrounded by the 
rosette-cells pass as fine threads down the funnel, and it 1s 
probably these that secrete the funnel, though the bases of the 
rosette-cells certainly reach the top of the funnel and may pass 
down it (fig. 39). 
As the stopper appears to be a definite portion of the micro- 
G g2 
