428 Bower.—Studies in the Phytogeny of the Filicates . 
The fertile leaves resemble the sterile, except that their pinnae are 
narrower, and their margins sharply recurved so as to protect the sporangia 
(Figs. 1 and 2). These are borne in large numbers upon the slightly 
enlarged receptacles which lie above the distal ends of the veins ; the sori 
are thus slightly elongated between midrib and margin. They seem to be 
always distinct laterally, and the insertion of sporangia on the leaf-surfaces 
between the veins, as described by Mettenius ( 1 . c., p. 5), has not been 
observed. 
On the same page of his memoir as that just quoted, Mettenius also 
describes for P. scandens ( 1 . c., Fig. 21) how the neighbouring secondary 
nerves of the pinnae are joined by intramarginal anastomoses. P. scandens 
from the Khasya hills has since been reduced to P. pycnophylla ; of this 
species I have material from Darjeeling and from Java ; in neither of these 
was any evidence found of such anastomoses. So far as my observations 
go the secondary veins are never united distally, and the sori which they 
bear are always laterally distinct from one another, except in those cases 
where there has been a basal forking of the vein. 
The roots are of the usual Fern type; they are not so numerous as in 
some other upright growing Ferns. Judging from the appearance in trans¬ 
verse sections, there are on the average more than two to each leaf; but 
they do not arise in any definite or constant numerical relation to them 
(Fig. 6). They spring from the central dictyostele, and usually arise from 
the margins of the foliar gaps. 
Anatomy. 
The anatomy of the genus Plagiogyria has received little attention 
hitherto. Mettenius ( 1 . c., p. 3) has recorded the fact that in P. semicordata 
a single vascular bundle enters each leaf from the stem ; this then divides 
into three as it passes into the swollen region of the leaf-base, but the 
It is to be noted that they are associated in these Ferns with a voluminous and slimy secretion 
completely covering the young circinate leaf; it arises from numerous branched hairs, the distal 
cells of which are enlarged as in Osmunda and Plagiogyria , and produce mucilage internally. It 
appears probable that there is a relation between the existence of the mucilaginous covering over 
the young part and the formation of the projecting pneumatophores; the latter are longer in the cases 
quoted than the thickness of the mucilaginous investment of the young part, and therefore project 
beyond its surface ; thus they would readily serve the purpose of gaseous interchange to the tissues 
within (compare Fig. 5). Moreover, their most important function appears to be past on the 
development of the leaf, for then they are apt to shrivel. From these facts the conclusion seems 
justified that they are provisions for the aeration, not of the mature parts, but of the young parts 
while covered by mucilage, and thus cut off from ready gaseous interchange with the atmosphere. 
It is to be noted, however, that they are not always present where even a thick mucilaginous felt 
exists; this is shown by the case of the Osmundaceae. From a morphological point of view they 
cannot be held as parts of high importance for purposes of comparison, since their inconstancy, 
even within a genus, indicates that they are opportunist growths, rather than permanent morphological 
features in any phylum. 
Organs similar to those described above have been noted by Kiihn (Flora, 1889, P* 486) in 
Nephrodiuvi stipellatum , Ilk., and an abstract of his results appears in Haberlandt’s Phys. Pflanzen- 
anat., 4. Aufi., p. 400, where the same function is attributed to them as suggested above. 
