54 Bower.—Studies in the Phytogeny of the Filicales. VIP 
Dennstaedtia , the fact that a vestigial lower indusium may sometimes be 
found would remove it. The small projection ( v.i .) at the lower limit of the 
receptacle in Fig. 40, a , is in the position of the lower indusium, and it 
is held to be its vestigial representative. It is shown under a higher power 
in Fig. 40, b. In most cases, however, the inner indusium is altogether 
absent, while the receptacle is often more compact and circular. In this 
condition the sorus appears superficial upon a vein, and distinctly intra¬ 
marginal. This is the characteristic of the Fern now designated in 
Christensens Index as Dryopteris (Pi)punctata (Thunbg.), C. Chr. It has 
been variously ascribed to Polypodium, Phegopteris , Hypolepis , N ephr odium, 
&c. Many authors have remarked how impossible it is to draw a definite 
line between it and species of Hypolepis} In particular Sir Joseph Hooker 
remarks on the genus Hypolepis in the ‘Flora Tasmaniae ’ 1 2 that ‘sometimes 
the reflexion of the pinnules’ margin is so slight that the sorus is really naked, 
and then I cannot distinguish the genus from Polypodium , or the species 
H. temdfolia from P. rugulosum , Lab.’ And later, referring to H. tenuifolia, 
he says: ‘ There is a New Zealand variety of this species with nearly 
glabrous rachis and stipes, more distant acuminate secondary and tertiary 
pinnae, narrow pinnules which are deeper lobed, and bear more numerous 
sori, scarcely covered by the involucre, and which hence passes into 
Polypodium rugulosum, Lab.’ Mr. H. Carse, comparing such types on the 
spot in New Zealand, remarks : 3 ‘In some forms of Hypolepis the spurious 
involucre (— inner, or lower indusium) is hardly or not at all developed, and 
sometimes it appears slightly in Polypodium , while the sori of the latter are 
frequently distinctly marginal.’ Thus he had already noted the vestigial 
indusium. 
The fact then appears to be that these Ferns are individually variable 
in the form and position of their sori. There is no obligation to draw any 
line between them. That they are very closely related is shown by their 
habit-similarity, by the absence of scales and presence of simple hairs, and 
by the practically identical vascular structure, as noted by Gwynne-Vaughan. 4 
We may take it that in this nearly related series, Hypolepis has been the 
designation of those forms which are more conservative of the Dennstaedtioid 
characters, while the more advanced forms have been referred to Phegopteris 
or Polypodium. 
A phyletic sequence may be traced thus. Starting from aDennstaedtioid- 
Dicksonioid source, with marginal sorus, a two-lipped indusium, receptacle 
conical, gradate sequence of sporangia, having oblique continuous annulus, 
the sorus becomes ‘ mixed ’, the receptacle flattened, and curved downwards ; 
1 Sir W. Hooker, Sp. Fil., vol. iv, p. 273 ; Syn. Fil., pp. 130, 312 ; Griesebach, Flora Brit. W. 
Ind., p. 67; Bentham, Flora Austr., vii, p. 765. 
2 Vol. ii, p. 138. 3 Trans. N.Z. Inst., vol, xlvii, p. 85. 
4 Ann. of Bot., vol. xvii, p. 694. 
