goa. . BOTANICAL GAZETTE [JUNE 
The figures also show that the cells are not all in the same plane, 
but are placed more or less obliquely above one another. This is 
not unlikely due to crowding by the large cell. Many such asym- 
metrical necks occur in my preparations. Occasionally one cell 
is so large as almost to pass for a central cell. These large neck 
cells possess large nuclei, as may be seen from fig. 50. Occasionally 
there is more than a single tier of cells in the neck; fig. 51 shows a 
neck of 3 tiers of cells. 
All the archegonial initials which I could certainly identify and. 
all young archegonia occur in the surface layer of cells. In one 
preparation there were 4 cells in a row. The outer one resembled 
a primary neck cell. The lowest one was large and had the general 
appearance of a central cell. The inner of the other two was about 
one-third as large as the lowest and slightly larger than the second 
one. An imbedded archegonium might have resulted, possibly, 
from such an initial as this. Though the mature archegonia are 
very frequently displaced and overgrown by the neighboring cells, 
it is ordinarily easy to find the free open passages from them to the 
surface. I did not find any case in which it was not very probable 
that such a passage exists. I am fairly convinced, therefore, that 
there are no deep-seated archegonia in this species of Araucaria. 
I believe that all such appearances are due to displacement and 
overgrowth. Eames has recently (3) expressed a somewhat similar 
opinion in regard to Agathis. I have examined a number of prepa- 
rations of A. imbricata and have seen no deep-seated archegonia 
among them. The displacement and overgrowth of the archegonia 
is rendered very easy Owing to the very delicate walls in the pro- 
thallus and to its frequently irregular outline. The more solid 
archegonium would thus be easily pushed into a position where the 
turgescent cells of the gametophyte would find an equilibrium of 
mutual pressures. The same fact would almost invariably lead to 
the crowding down into the archegonial cavity of the adjacent cells 
if they encountered any resistance at all in the expansion of the 
cavity in which the gametophyte grows. 
The jacket usually consists of a single layer (figs. 42, 43) of more 
dense cells, which are usually uninucleate and with comparatively 
thin walls. The wall next the egg is somewhat thicker than the 
