1910] MCCUBBIN—DEVELOPMENT OF THE HELVELLINEAE 203 
the same phenomenon in Mitrula phalloides. It is evident that 
in these cases the developing ascus must be nourished by diffusion 
from the surrounding cells. As growth proceeds in the vertical 
branch, transverse walls are formed at intervals along its course 
(figs. 11-18). When nuclei and protoplasm happen to be cut off 
by these walls, a lateral branch may be produced from the cells thus 
left behind (figs. 12, 16). Finally the end cell forms a hook and its 
two nuclei divide to form four. Then follows a septation of the hook 
into three cells, a daughter of each of the mother nuclei remaining 
in the penultimate cell and the other two being distributed to the 
terminal and antepenultimate respectively. From the penultimate cell 
a process now arises into which its two nuclei wander (figs. 20, 27.) 
Heretofore in the Ascomycetes which have been investigated there 
has come into consideration only one hook (where hooks exist), 
the process from which becomes the ascus. Here, however, it is 
curious to note that the process from the first hook may proceed to 
form a second. Its two nuclei, instead of fusing, divide again, the 
resulting four being distributed in the second hook as in the first. 
Indeed this formation of successive hooks may go on for some time, 
as many as six having been observed on one ascogenous hypha. The 
last-formed process then becomes the ascus, its nuclei fusing to form 
the primary ascus nucleus, and from this spore nuclei are produced 
by the usual three successive divisions. One important feature is 
made strikingly evident by these hooks, namely the conjugate division 
of the nuclei in the ascogenous hyphae. It has been quite evident 
from my observations that the two nuclei which fuse in the young 
ascus are directly descended by conjugate division from the two 
which normally occupy the end cell of the ascogenous hypha as it 
first arises from the subhymenial layer. 
Another phenomenon of interest is the “fusion” or “anastomosis” 
that quite frequently takes place between the terminal and ante- 
Penultimate cells. The nucleus of the antepenultimate cell passes 
into the former, and from thence the two nuclei wander into a process 
Which invariably arises from the terminal cell (figs. 33, 34, 36-38, 
40). It will be readily seen that this process with its two nuclei is 
the direct equivalent of that which originates from the penultimate 
cell, and indeed it conducts itself in exactly the same way, forming 
