162 Proceedings of the Royal Society of Edinburgh. [Sess. 
the action of the vitellophags. Each mass of yolk, which contains in its 
centre a vitellophag with one or two nuclei, eventually becomes limited by 
a cell-boundary and forms a yolk-sphere (see also fig. 4). Hitherto the 
yolk-cells have multiplied by indirect division, but after the formation of 
the yolk-balls they exhibit no more mitotic figures. 
Meantime the germ-rudiment itself has become crescentic in outline, 
due, as a reference to fig. 8 6 will show, to its having accomplished a turning 
movement, which brings its opening to face laterally. By the continued 
inward growth of its edges it has become pouch-like, and its lateral aspects 
are now very much wider than the dorsal and ventral (see fig. 8c). 
I have not met with any allusion to a similar turning movement in 
accounts of the development of the germ-rudiment in other Lepidoptera, 
which, however, as stated above, are always meagre. Probably the turning 
is much more marked in Eudemis than in other Lepidoptera hitherto in- 
vestigated, owing to the very flattened form of the egg, the rotation being- 
necessary in order to permit the subsequent growth of the appendages. 
An observer of a living egg in the state shown in figs. 7 a and 7b is unable 
to predict whether the germ-rudiment will turn to the right or left. 
Probably the direction is determined by the origin of the egg from the 
right or left ovary,* and is in accordance with “ the law of orientation ” of 
Hallez, which affirms that the cephalic pole of the egg within the ovary 
is directed towards the head of the mother, and that its ventral and dorsal 
aspects are also coincidently fixed, consequently also its right and left sides. 
A female Eudemis kept in a small breeding cage with a glass front will 
deposit many eggs on the glass, and when development has sufficiently 
progressed it can be clearly seen that while many of the eggs contain 
embryos curved to the right, the rest are curved in the opposite direction. 
Further, a large number of eggs collected in the open, removed from the 
leaves and examined, showed that the right and left curvatures were about 
equally represented. It must not be forgotten that the ultimate curvature 
of the fully formed larva in the egg will be in the opposite direction to 
that of the germ -rudiment at the stage under discussion, owing to the 
subsequent greater growth of the dorsal tissues. 
A comparison of the transverse sections shown in figs. 6c and 8c will 
show that the tissue of the germ-rudiment, though still consisting of only 
one layer of cells, has now about doubled in thickness, and the altered 
form of the cells may be gathered by a comparison of figs. 5a and 56. 
* Whether an egg from the right ovary produces a dextrally or a sinistrally turned germ- 
rudiment has not, so far as I am aware, been ascertained. A study of the curvature of 
the embryo in certain viviparous insects might throw light on this matter. 
