102 TuHompson: THE MORPHOLOGY OF TAENIOMA 
both the Porto Rican plant with the three-haired branches, and 
the Bahamian one with the two-haired branches produced tetra- 
sporic stichidia. Agardh (’63) speaks of the stichidia as “ often 
excurrent at the top into three minute hairs, of which the middle 
one continues the middle part of the segment [7. ¢., the midrib] 
the marginal ones the marginal cells” [translation]. This cor- 
responds exactly to my Porto Rican material. There the tetra- 
sporangia are formed in stichidia on the flat shoots of the secon- 
dary axes in considerable number (Fic. 12). These stichidia are 
flat branches which remain practically unmodified except that, 
with the formation of the large tetraspores, they become some- 
what broader. The tetrasporangia are formed in two long rows 
from the two pericentral cells which lie one on each side of the 
midrib, A transverse division first separates these cells into an 
upper and a lower half (Fic. 13). The upper half becomes the 
mother-cell of the tetraspores. The lower half divides by two 
longitudinal divisions parallel to the flat surface of the branch into 
a layer of three cells in thickness. These become crowded and 
elongated in a radial direction by the growth of the tetraspore 
mother-cells. The upper half of the pericentral cell enlarges 
greatly to form the tetraspore mother-cell, which still remains in 
protoplasmic communication with the lower layer of supporting 
cells. To form the tetraspores, the contents of the mother-cell 
divide first into two by a longitudinal wall parallel to the flat sur- 
face of the stichidium (ric. 14). A transverse division forms two 
tetraspores from one half, and another division, at right angles to 
the planes of the first two divisions, correspondingly forms the two 
in the second half. These are therefore one form of “ cruciate”’ 
tetraspores. They are held within the mother-cell wall, which 
breaks when mature, allowing the tetraspores to escape. These 
lateral pericentral cells are the only ones to be modified, the midrib 
and the marginal cells being unchanged. The entire branch or 
only a portion of it may be transformed into a stichidium. 
Little has been known heretofore concerning the antheridia. 
Falkenberg (’oI) figures a shoot bearing two large conical pro- 
tuberances, seemingly filled with, or composed of, small cells, but 
he rightly concludes that these are malformations and not an- 
theridia. De-Toni (’00) mentions antheridia as spots between the 
