104 THE SPORANGIUM DEFINED 
of the spores. The extent of the wall as such is closely related to the 
position of the sporogenous cells: where these are deeply sunk, the wall 
is a mere roof over them: where they are carried outwards by the growth 
of the tissues surrounding them, and a projecting sporangium is formed, there 
the wall envelopes them as a tissue of greater extent. But in either case 
it is continuous with the tissue of the sporangium-bearing part, of which it 
is to be regarded as a specially developed region. This view of it accords 
well with the structure of sunken sporangia, as in Ofsvoglossum (Fig. 59), 
in which the tissues of the wall are continuous with and little differentiated 
from the cognate tissues of the spike. The same is the case with other 
Ophioglossaceae, even in those where the sporangia project ; but in these, 
as also in the Lycopods and Equiseta, there is some further specialisation 
of the wall for dehiscence than 
in the deeper-seated forms. In the 
Filicineae still more exact specialisa- 
tion is the rule, and the mechanical 
annulus becomes a marked feature 
in the stalked sporangium. But a 
comparison of the Marattiaceae and 
other Ferns leads to the conclusion 
that the presence of an annulus, 
and its elaborateness, are to be 
correlated with the freedom of the 
sporangium from mutual relations 
with other bodies. The annulus, 
in fact, is still only a specialised 
region of the sporangial wall. This 
conclusion accords with the facts 
Fic. 59. of its inconstancy, for sporangia 
eee a 
shaded. "x 100. : no such mechanism. That is the 
case in J/svetes, while the sporangia 
of the Hydropterideae are also entirely without an annulus; nor is there 
any stomium in them, which would localise dehiscence. Loxsoma is a 
specially interesting case, for there the annulus, though indicated by the 
cell-divisions as complete, is only partially indurated: as a matter of fact, 
the part of it which is not indurated could not possibly be mechanically 
effective, owing to the mode of packing of the sporangia in the sorus 
(Fig. 60). 
The general conclusion to be drawn is, that while the protective wall 
itself is always present, those mechanical arrangements collectively desig- 
nated by the term “annulus” are not essential or constant parts of the 
sporangium, however constant they may seem to be in certain groups: 
where they cannot be mechanically effective they may be entirely omitted. 
Nor is the ¢afetim to be looked upon generally as a morphological 
