THompson: THE MORPHOLOGY OF TAENIOMA 103 
middle vein and the margins of the branches, a description which 
does not wholly correspond to the antheridia as seen in the 
material at hand. In this, antheridia are formed, as are the tetra- . 
sporangia, in ordinary flattened branches, clustered near the grow- 
ing tips of the secondary axes (FIG. 15). The entire branch (with 
the exceptions noted below) is usually transformed into antheridial 
cells, although these may be found only in the middle portion, 
while the cells at base and apex of the branch remain unchanged. 
In the formation of the antheridia the first divisions occur in the 
row of marginal cells. A longitudinal split cuts each into two 
(Fic. 17). After the first division of the marginal cells into 
two, the outer row of these never divides again, so that, in the 
mature antheridium, there is always to be seen a row of marginal 
cells one half the size of the original ones (Fic. 18). The inner 
row divides irregularly into many small cells. The lateral peri- 
central cells now become modified, being cut by both longitudinal 
and transverse divisions into numerous small cubical cells (FIG. 16). 
Before these are entirely completed, a longitudinal split parallel to 
the flat surface of the shoot takes place. This causes the shoot 
to become a flattened plate of small angular cells, two layers of 
cells in thickness, with two exceptions: (1) the midrib, which re- 
mains entirely unchanged, and (2) the marginal cells. Next, the 
true antheridial cells are separated from these mother-cells in the 
two layers on the side toward the surface of the flat shoot (FIG. 19). 
These very small, ovoid bodies, containing the spermatia, are all 
in protoplasmic connection and are formed 2-4 toa mother-cell, 
4 being probably the normal number. At regular intervals in the 
antheridial branch there can be noticed a faint line of separation 
from the midrib to the margin, denoting the original separation of 
cells into segments (FIG. 15). 
Although the tetraspores have been known since the plant was 
first described, and the presence of antheridia has more recently 
been alluded to, the cystocarp has been absolutely unknown. De- 
Toni (’00) says specifically “ Cystocarpia ignota”’ and other liter- 
ature on the subject does not mention them at all. It has been 
my good fortune to find in the material collected at Porto Rico a 
number of cystocarpic specimens. Unfortunately the material has 
not afforded sufficient stages to trace the growth of the cystocarp 
from its beginning, nor has there been a sufficient number of 
