296 Phillips. — On the Develop 7 nent of 
which belong to the auxiliary cell, that with the central cell, 
and those with all three of its branches, remain with the 
inferior cell. The auxiliary cell has in fact given rise to a new 
branch consisting of one cell, and it is this derived cell which 
Schmitz in his later account considers the true auxiliary cell. 
From it the sporogenous filaments — meta-ooblastemata as 
Schmitz has called them — arise. These meta-ooblastemata 
branch freely in a sub-dichotomous fashion, and form the 
close-set mass of filaments known formerly as the ‘ nucleus,’ 
the terminal cells of which ultimately discharge the densely- 
coloured pyriform carpospores. 
The effect of the formation of the ‘ nucleus ’ upon the cells 
of the carpogonial branch is to detach them from their pit- 
connexion with the auxiliary cell, and to push them out- 
ward towards the pore, where, thus shunted, they may still 
be detected at a late stage as a row of shrunken cells 
(Figs. 4, 6 , 14). 
The second effect of fertilization is that upon the two sterile 
branches. The inferior one-celled branch adds a new cell, 
while each of the two cells of the two-celled branch buds off 
a new cell ; so that the sterile products of the auxiliary cell 
now number six cells. Beyond this the development does 
not go ; but as the cystocarp becomes filled with carpospores 
it becomes difficult to trace the changes in the sterile tissue. 
The six sterile cells may still be seen at a late stage of spore- 
formation (Fig. 4). 
A third effect of fertilization is the branching of the central 
cell itself laterally, into two cells which branch again and again, 
forming at length about twelve filaments which line the whole 
cavity of the cystocarp extending as far as the pore (Fig. 9). 
The wall of the cystocarp is properly only one cell thick, but 
the existence of this layer gives it the appearance of being 
two cells thick. The cells of the lining do not, however, form 
a pseudo-parenchyma as do those of the pericarp, but retain 
their filamentous character to the end, separated laterally by 
a wide interval, circumstances which render them easy of 
identification. These cells are no doubt the paranemata of 
