of Pilularia globulif era, L. 257 
the primary members of the embryo, we may say that the 
second leaf originates like the later ones, as a segment of the 
apical cell of the stem. Like the stem, it grows slowly at 
first, and in case the apical cell persists, is scarcely to be 
distinguished from it (PL XIV, Fig. 22 d, l 2 ). About the 
eighth day, however, it begins to elongate, though much more 
slowly than the primary leaf, and from this time onwards 
is very easily recognised (PL XIV, Fig. 25 ; PL XV, Figs. 
1, 8). About two weeks after sowing the spores the second 
leaf begins to grow rapidly, and in a very few days reaches 
its full size. 
The Foot-quadrant. 
The first divisions in the foot-quadrant (PL XIV, Figs. 8, 9, 
&c.) follow closely those of the root, but this regularity soon 
ceases, and after the first two or three divisions no definite 
succession of the walls can be distinguished. The foot never 
attains any great size, and as already said, all the lower cells 
of the embryo probably absorb the nutriment from the spore. 
As the embryo grows the prothallium keeps pace with it for 
some time. About the time the embryo is divided into eight 
cells, the upper part of the archegonium has its cells divided 
by tangential walls, so that this part of the embryo is sur- 
rounded by a double layer of cells (PL XIV, Figs. 10, 13). 
Both Hofmeister and Arcangeli 1 figure this condition in the 
unfertilized archegonium, which in Pilularia never occurs. 
The basal cells of the prothallium divide further and develop 
numerous root-hairs. The plasma in the upper part of the 
spore increases in quantity as the embryo develops and pushes 
up the base of the prothallium and embryo, which become in 
consequence strongly concave below (Pl. XIV, Fig. 23). Al- 
though the nucleus was not observed in actual division, in a 
number of instances in the later stages of development bodies 
which behaved with reference to staining agents in the same 
way as nuclei were seen in this plasma-mass, and were quite 
probably derivations of the original c endosperm nucleus.’ 
1 1. c., Plate VIII, Fig. 5 . 
