vm و‎ i GES ben, BS 
DEVELOPMENT 21 
which it is found. This cylindrical cell grows for a time in all 
dimensions, but before long begins to enlarge greatly at the upper 
end where the protoplasm of the entire spore will soon be found. 
As soon as this migration of protoplasm from below is completed, 
a partition wall is formed, separating this now ellipsoid upper por- 
tion from the lower cylindrical one. This process results in the 
formation of a single fertile cell with delicate hyaline homogeneous 
appendage at its lower end. This constitutes the primary appen- 
dage spoken of throughout this paper, and is really an abortive 
cell or possibly only an abortive portion of a cell; the determin- 
ing of this point would necessarily involve the behavior of the 
nucleus. One may often see on rupturing the perithecium that 
the wall laid down between the fertile and abortive cells is very 
delicate compared with the remainder of the spore boundary, for 
the slightest disturbance often ruptures it and allows the proto- 
plasm to flow out of the upper cell into the now empty lower one. 
This phenomenon can be seen very readily in such forms as 
P. decipiens, P. vestita and P. fimiseda. The same thing is shown 
in a capital manner in P. zygospora. Here the connecting fila- 
ment between the two fertile cells of the spore is of the same 
nature as the primary appendage in the other species ; but this 
species deserves special treatment on account of its unusual form. 
Here the young spores are approximately crescent-shaped, and 
their contents resemble that of the ascus as before. A tremen- 
dous increase in length takes place before there is any appreciable 
terminal enlargement. About the time that this has begun to 
manifest itself, the filaments themselves are comparatively longer 
than the spore-bearing portion of the ascus ; and have, therefore, 
become twisted in their growth so as to display a spiral arrange- 
ment within the ascus. When enlargements begin they appear at 
each end of this long filament instead of at one end of a compara- 
tively short one as in the other species of the genus ; and the pro- 
toplasm from the filament moves toward either end, forming two 
fertile cells instead of one as in other species. This results in the 
production of sixteen fertile cells within the ascus. Inasmuch as the 
filament joining two cells soon dissolves, the ascus appears to be, 
and is in effect a 16-spored ascus. On this account Saccardo has 
transferred the species to the genus Philocopra. When the method 
