516 FILICALES 
round numbers 60, spore-mother-cells traversed. Tangential sections show, 
on the average, about six such layers of cells to be present, and the 
average number of spore-mother-cells in each sporangium would thus be 
6 x 60 = 360, while the average potential output of spores per sporangium 
would be 360x4=1440 spores. It may be noted that, in the cases 
observed, all the spore-mother-cells undergo the tetrad-division, and none 
have been seen to be disorganised, though this, doubtless, may occur 
occasionally. 
Reverting to the tissues outside the sporogenous group, the tapetum is 
not a strictly defined layer, and is often irregularly doubled by periclinal 
divisions ; but this is by no means always the case. It remains recognis- 
able as distinct, enlarged, glandular cells, often with several nuclei, up to 
the period of the tetrad-division. The wall of the sporangium outside the 
tapetum consists of two or more layers, commonly three, of which the 
outermost is the firmest and most differentiated; the inner layers are less 
regular, and are composed of more or less tabular cells. I have never 
seen the tapetum in direct juxtaposition with the superficial layer, as it 
has been figured and described by Campbell.! Of the external wall three 
essential parts may be recognised, and they will be best seen in Fig. 284 pb, 
which represents the apex of the sporangium, cut off in such a plane 
as x, x in Fig. 284 8. The first part (a, Fig. 284 £) is found on the 
peripheral side of the sporangium, and consists of large turgid cells, 
with moderately thin walls, and granular contents, which stain deeply 
with Bismarck brown. A second region (4) consists of deep prismatic 
cells with thick lignified walls, which may be recognised as the annulus; 
it extends as a narrow bridge across the apex of the sporangium (Fig. 284 D), 
and widens out on either side, as the apex is left, into a broader band 
of cells with lignified walls (Fig. 284 F). The third region (c) consists 
of thinner-walled cells, of elongated form, which constitute a narrow 
band running down the anterior (ventral) side of the sporangium. This 
is the tissue which defines the fissure of dehiscence. 
The structure of the sporangial wall being as thus described, we 
may now consider how it works in connection with dehiscence. The 
annulus, together with the two broader lateral extensions of it, constitute 
a firm resistant arch, of which the apex is the narrowest part, being 
only about three cells wide (Figs. 284 D, £). If the thinner-walled posterior 
region (a) were to contract, as we may presume it does by drying as the 
sporangium matures, the two sides would be pulled backwards, while the 
thin bridge of the annulus at the apex would act as a sort of semi-rigid 
hinge; the line of dehiscence on the ventral face, having been structurally 
defined, would thus, on fission, be caused to gape widely. It is not 
probable, however, that this hinge-like action is very considerable, and 
the gaping of the slit may be mainly due, as in other Marattiaceae, to 
mere drying up of the cells in the neighbourhood of the rupture. 
1 Mosses and Ferns, p. 297, and Fig. 164 c. 
