440 Bower.—Studies in the Phytogeny of the Filicales. 
their sequence, while occasionally single cells of the sporangial wall quite 
distinct from the annulus may be indurated. But putting these irregularities 
aside, the annulus of about thirty cells takes an oblique course round the 
sporangium, being quite continuous past the insertion of the stalk. The 
obliquely distal area of the sporangial wall surrounded by the annulus, 
seen in surface view in Fig. 18, clearly corresponds to the ‘peripheral ’ 
face in the Gleicheniaceous type, and is here, as in Gleichenia y only slightly 
convex ; while the rest of the thinner part of the wall to which the stalk is 
attached is strongly convex, and corresponds to the ‘ central 5 face of 
Gleichenia (Fig. 18, a). A similar comparison may be drawn also with 
Dicksonia. Comparing with the Schizaeaceae, the differences appear at 
first sight more striking than the resemblances, for in most of them the 
* peripheral ’ face is greatly reduced; but it has been seen that the nearest 
anatomical correspondence is with Anemia , and in this genus the ‘ peripheral ’ 
face is larger than in any other genus of the Schizaeaceae ; in fact, notwith¬ 
standing its short stalk, its apparently transverse annulus, and its peculiar 
form, the type of sporangium is essentially the same in Anemia as in 
Ptagiogyria. It is not desirable to press such comparisons too much into 
detail but it is worthy of remark that on the side of the annulus nearest to 
the stalk in the sporangia shown in Fig. 18, b and the series of cells is 
duplicated ; this corresponds exactly in position to that extension of the 
annulus down the side of the sporangium in Anemia which is related to the 
function of dehiscence (compare Engler and Prantl, i, 4, Fig. 192, g). 
Turning to the mechanism of dehiscence as seen in Ptagiogyria , this leads 
to the formation of a lateral slit, as in Dicksonia and most other Gradatae; 
but the structural provision for it shows some variability, which may be 
held to suggest the sort of changes which would be involved in a shift¬ 
ing of position of the dehiscence. In the sporangia of the Simplices there 
is no highly organized ‘stomium’; and this applies equally for Anemia ,' 
with which comparison has above been drawn. In Ptagiogyria the dehis¬ 
cence is localized laterally by a well-organized mechanism ; but a comparison 
of various sporangia of this plant shows that there is uniformity neither 
in the exact position nor in the size or number of the cells which determine 
it. There is usually a well-defined group of four sister cells, which con¬ 
stitute the actual stomium (Fig. 18, a, b)\ on either side of it (sometimes 
in immediate juxtaposition with these, or sometimes not) is a group of thin- 
walled cells, which, however, vary in number ; thus on the distal side of the 
stomium there may be only one of these cells (Fig. 18, d), or two (Fig. 18 ,c), 
or three (Fig. 18, b) ; on the proximal side there may be two (Fig. 18,4), 
three (Fig. 18, <2), or four (Fig. 18, d ), while they may be in direct juxta¬ 
position with the group of cells of the stomium (Fig. 18, a), or removed 
from them on either side by the intervention of a variable number of 
indurated cells (Fig. 18, b, e). This variability, so different from what 
