1904] HOLFERTY: ARCHEGONIUM OF MNIUM 113 
THE DIVISION OF THE CENTRAL CELL. 
The division of the central cell results in many cases in two 
cells of nearly equal size. This is well illustrated in figs. 26-29, 
in which the ventral canal cell is nearly as large as the egg. 
This character strengthens the view that the difference in size 
and function of the egg and ventral canal cell may not always 
have existed. In fact, an equality in size and probably in func- 
tion may reasonably be regarded as a primitive condition; and 
the existing inequality may be regarded as a specialization, which 
among pteridophytes and gymnosperms has led to a gradual 
reduction in the relative size of the ventral canal cell, then to 
the disappearance of the wall separating its nucleus from that of 
the egg, and finally to the entire disappearance of any vestige 
of the cell. Goebel (18, p. 242) refers to this approximate 
equality in size of egg and ventral canal cell in the bryophytes, 
saying that they are nearly equal in many cases, but that in most 
cases the egg is larger. He admits that the meaning of this 
division is not known, but since the ventral canal cell is always 
present, he is inclined to believe that it has some physiological 
role, and suggests the separation of the chemotactic substance 
for the attraction of the sperms as a probable function. 
The division of the central cell is comparatively late. No 
case was found where division occurred before a row of seven 
cells had been formed (fig. 34), and very often eight or ten 
(figs. 32, 33). No mitotic figures for this division were found, 
but the position of the nuclei and the septum, and other features, 
are such as to leave no doubt whatever concerning the relations 
of these two cells. A wall separating egg and ventral canal is 
always present for some time after the division takes place, but 
not infrequently it disappeared while the remaining septa of the 
row were still intact. 
TRANSVERSE SECTIONS OF THE ARCHEGONIUM. 
All the evidence for the development of the archegonium 
thus far presented has been drawn from longitudinal sections. 
Several series of cross-sections were made, two of which are pre- 
sented herewith, and from them the following conclusions are 
