380 Bower. — ■ The comparative examination of the 
tinct divisions of the Ferns ; viz. in the Hymenophyllaceae in 
which it is probably the typical structure ; in species of 
Asplenium and of Todea , in which it is rather to be regarded 
as the result of a retrogressive adaptation. 
An attempt has also been made in the above pages to make 
a comparative study of those wings which are so commonly 
found on the leaves of Ferns. It has generally been assumed 
that because those wing-structures which are continued, or 
decurrent from the leaf to the stem, are of irregular occurrence, 
therefore wings, whether on stem or on leaf, are of little 
morphological moment. This is an entirely fallacious con- 
clusion. It is a matter of common observation that such wings 
are very constant in appearance and position on leaves of large 
classes of plants, more especially of the Ferns and Dicotyledons. 
Now one of the most important factors in stamping the mor- 
phological importance of any part is its constancy of occur- 
rence and character in considerable series of organisms, and 
on this ground the wings deserve special attention. It may 
be objected, however, that wings are merely a natural conse- 
quence of the flattened development of leaves, and that an 
elongated and flattened structure must have two marginal 
lines of greatest curvature. This is true, but it does not 
diminish the interest of tracing those margins, comparing 
them in different plants, and noting their various modes of 
development in the different parts of the same leaf. 
I have repeatedly dwelt upon the fact that though the 
leaves of the Filicineae may develop occasionally in a cylin- 
drical form ( Pilidaria ), or as spathulate structures without 
a midrib ( Drymoglossum ), those of the vast majority of them 
have a robust midrib, which is traversed longitudinally by two 
more or less developed lateral flanges or wings. These ori- 
ginate in the lowest Ferns from definite parts of the segments 
cut off from the apical cell, while in the higher, though their 
position is the same, the segmentation is less definite. In the 
mature state they usually show more robust and massive 
characters in the lower part of the leaf. This has been 
recognised both in external form, and in the internal struc- 
