meris terns of Ferns as a Phylogenetic Study. 381 
ture in Trichomanes reniforme , in Scolopendrium vulgar Aspi- 
dium Filix mas , Asplenium nidus , P ter is aquilina , and may 
be verified in most Ferns: more especially however is it to be 
noted in the higher Ferns of the series ; and I have elsewhere 
pointed out 1 that the bulky developments which are found at 
the base of the leaf of Osmunda and Todea> and even the 
■ stipular’ growths of the Marattiaceae are to be regarded as 
the result of metamorphosis, or special conformation of these 
wings, or of the extension of the growth across the face of the 
phyllopodium, so as to form the well-known commissure. 
There is thus a certain relation between the greater bulk of 
the base of the leaf, and that of the wings which it bears ; 
but the wings of the lower part, though of bulky development, 
are but little expanded in area. In that part of the phyllo- 
podium which is usually distinguished as the petiole, the wings 
are of a reduced type, and are seen merely as pale streaks, 
often discontinuous, which run down the sides of the petiole, 
and, though they may not project at all, these streaks show 
where the morphological margin of the leaf lies ; sometimes, 
however, the wings project even here as distinct flanges (many 
Hymenophyllaceae). It is, however, in the upper part of the 
leaf that the wings attain their typical development, and there 
they constitute the large proportion of the expanded surface, 
whether of the simple or the branched leaf. Thus the wings are 
subject to great differences of development in different parts 
of the leaf, and indeed the general conformation of the dif- 
ferent parts of the leaf depends more closely upon this than 
upon differences of development of the thickened midrib. It 
is also a matter of common experience that from them arise 
the branchings of the leaf. 
It will be recognised that this mode of viewing the leaf 
throughout its length as a whole, and avoiding at the outset 
those horizontal distinctions into ‘ foliar base ’ and £ upper 
leaf,’ which are habitual with other writers, naturally leads to 
a more coherent view of the real nature of the leaf. It is 
typically a flattened structure, of which the longitudinal lines 
1 Phil. Trans. 1884, Part II, pp. 578, etc. 
