1922] DUPLER—REBOULIA 153 
cites a case where the basal cell has divided into a small group of 
irregular cells, probably also a very rare feature. 
It is evident that the basal cell of the filamentous embryo under- 
goes but few, and in some cases no further divisions, and therefore 
makes a relatively small contribution to the tissue of the sporophyte, 
which therefore is built up almost entirely from the three anterior 
cells of the filament of four cells. 
LATER GROWTH AND EMBRYO DIFFERENTIATION.—Along with or 
following the formation of vertical walls, transverse divisions in 
some or all of the segments result in additional tiers of cells, the divi- 
sions occurring in different planes without any definite sequence 
(figs. 36-41). At first the ventral side of the embryo will probably 
show the greater number of cells (figs. 35, 39, 40), but as growth 
continues the dorsal side also grows rapidly, and the embryo soon 
becomes a radial instead of a bilateral structure (fig. 46). Periclinal 
walls now form, especially in the epibasal portion of the embryo, 
without any definite sequence, forming inner and outer cells (figs. 
38-44). These first periclinal walls are most likely to form in the 
capsule-forming region of the embryo, which becomes considerably 
broader than the more slender hypobasal portion (figs. 40-45). 
This portion, however, soon broadens out somewhat and reaches the 
winter condition (fig. 46). While it is impossible to trace back ab- 
solutely the origin of the different regions of the sporophyte, it 
seems most probable from the writer’s study that the first division 
of the egg determines the capsule region as distinguished from the 
foot and stalk region, the epibasal cell giving rise to the capsule, 
therefore, the hypobasal to the stalk and foot, with the bulk of both 
foot and stalk derived from the anterior half or three-fourths of the 
original hypobasal cell. They are derived from the anterior half if: 
the basal cell (of the row of four) does not undergo transverse divi- 
sion, and from the anterior three-fourths if the basal cell should 
divide transversely. Both Woopspurn and Haupt regard the foot 
as derived from the hypobasal cell, the stalk and capsule from the 
epibasal cell; the capsule being formed from the two anterior cells 
of the three derived from the epibasal cell, according to Haupt. 
The sequence of the early divisions and the behavior of the hypo- 
