JUNGERMANNIALES 265 
the basis of the first segmentations, of the capsule from the seta. So far 
as segmentations offer a basis for argument, the general conclusion may 
be drawn that the seta and capsule are not always distinct ad ‘vito. The 
developmental facts suggest that the fertile region may be held to be a 
residuum left by sterilisation, which has been basifugally progressive : the 
result of such sterilisation is the region of the seta as it is seen in these 
plants. We may regard as the most primi- 
tive case that in which all the tiers of cells 
of the embryo form .the archesporium : 
those cases in which the differentiation 
of the archesporium is deferred in the 
individual life may be held to be relatively 
less primitive. 
The four cells of the uppermost tier, 
which thus as a rule form the capsule 
in the Jungermanniales, divide first 
transversely to form four terminal cover- 
cells: the four larger cells below again 
segment to form four inner cells and 
eight to twelve peripheral cells. This 
is the same segmentation as occurs also 
in the lower tiers: the similarity is in 
accordance with the view above expressed 
as to the origin of the seta by basifugal 
sterilisation, and supports the conclusion 
that seta and capsule had a common 
origin. 
However interesting such questions 
may be, they are more or less speculative. 
A much greater interest, proportional to 
the greater cogency of the facts, attaches 
to the various modes of development of 
the capsule itself in the Jungermanniales. Fic, 126. 
It has been seen that the inner cells Porella Bolanderi. Longitudinal section of 
above described constitute the arche- ee se era ig 
sporium. In many of the Akrogynous 
Jungermanniaceae the cells, after repeated divisions, undergo a differen- 
tiation as in the Marchantiaceae, into spore-mother-cells and sterile elaters : 
these are associated in various ways, and the case of Foredla will serve 
as an average example (Fig. 126). The argument from differentiation of 
sterile and fertile cells during development applies equally here as in the 
Marchantiaceae. The same is the case with many of the Anakrogynae ; 
but in some of the latter there is a more specialised tissue-differentiation 
leading to the formation of a coherent mass of sterile tissue, with a more 
definitely localised residuum of fertile tissue: this sterile mass has been 
