i6o 
AMERICAN JOURNAL OF BOTANY 
[Vol. io, 
the simpler Marchantiales. A possible basis for such an assumption is 
afforded by Marchantia (Durand), in which the four neck-canal cells some¬ 
times become binucleate as though preparing for this suppressed division; 
but this fact might also be interpreted in a converse manner to support the 
assumption that the condition in the Sphaerocarpales is the more primitive 
one, and that the Jungermanniales and the higher Marchantiales have 
acquired the habit of additional divisions in the neck-canal cells. It has 
not been shown whether or not environmental conditions may affect the 
number of divisions in the neck-canal cells. Hutchinson found only half 
as many neck-canal cells in Pellia epiphylla as did Janczewski. This 
reduction may conceivably have resulted from a suppression of the last 
division due to the influence of the environment in which his plants had 
grown. 
Non-functional Archegonia 
Abnormal sex organs are generally recognized as occurring frequently 
in the Bryophyta, and Florin (1918) has described several abnormal arche¬ 
gonia in Riccardia pinguis. One of these is an archegonium with a single 
egg and a double canal row. In my material I have found one archegonium 
which may be considered a complementary case. In this archegonium there 
are two eggs and two ventral canal cells, but only one row of neck-canal 
cells (fig. 38, PI. XVIII). One egg and one ventral canal cell form the usual 
part of an apparently normal axial row. The additional egg and ventral 
canal cell appear slightly lower down in the archegonium and are separated 
from thctee of the complete axial row by a film which seems to be a cell wall. 
This archegonium is considerably more massive than those of the usual type, 
and a rift appears between the wall proper and the cells immediately above 
the supernumerary ventral canal cell. I have not been able to determine 
how far this rift extends around the column of cells above the ventral canal 
cell. The cells of this column have angular walls and numerous plastids and 
present the appearance characteristic of cells of the archegonial wall, with 
which the column is continuous above. 
In the locations from which my material was collected Riccardia pro¬ 
duces large numbers of archegonia, of which only a small percentage function 
in reproduction. It is apparent from the sections that a large percentage 
of the mature archegonia in the material collected were not capable of so 
functioning. In many cases it appears that disintegration has begun in the 
egg and the ventral canal cell before the maturity and normal disintegration 
of the neck-canal cells (figs. 36, 37, PI. XVIII; figs. 42-46, PI. XIX). Those 
in which this disintegration appears to be well advanced stain deeply with 
safranin although the cytoplasm of the neck-canal cells does not take this 
stain (figs. 36, 37, PI. XVIII; and figs. 42, 45, 46, PI. XIX). This premature 
disintegration of the egg and ventral canal cell is generally accompanied 
by a marked shrinkage of these cells, and the later swelling of the neck-canal 
cells sometimes causes the latter to push down into the venter (fig. 44). 
