108 Davie.- — The Development of the Somes and Sporangium 
cuticle which covers the lid-cell persist as a fringe round the neck of the 
antheridium after the lid-cell has been discharged. The resemblance to 
the ruptured antheridium of the Polypodiaceae is then so close as to 
destroy the value of the antheridial dehiscence-method as a differential 
criterion (Schlumberger, Tl, p. 396). In Peranema the lid-cell of the 
antheridium is generally undivided and is discharged at' the maturity of 
the antheridium, which then presents exactly the appearance of the an- 
theridium of Woodsia ilvensis , figured by Schlumberger (loc. cit., Figs. 7 
and 8). Only one antheridium was found with a divided lid-cell (PI. Ill, 
Fig. 8), though many were examined in the youngest, semi-mature, and 
mature stages. This feature of one antheridium, taken along with the 
features of some of the glandular hairs, mentioned above, shows that 
Peranema retains a suggestion of the Cyatheaceous type in its prothallus, 
which on the whole approaches even more closely than that of Woodsia 
ilvensis to the type of the Polypodiaceae. 
Among the three closely related genera within the Woodsieae- 
Woodsiinae group, to which Peranema undoubtedly belongs, the sequence 
now appears to be Woodsia — Diacalpe — Peranema. Woodsia has a Gradate 
sorus, and prothallial features which relate it to the Cyatheaceae ; Dia- 
calpe has a mixed sorus, with a basal indusium, and an unmistakable 
Cyatheoid prothallus, which give it an intermediate position ; Peranema 
has a mixed sorus, with a modified basal indusium and a Nephrodioid 
prothallus, which place it near to the Aspidieae. Already in the features 
of the sori and prothalli of Woodsia and Diacalpe we have had indica- 
tions of a line pointing towards the Polypodiaceae, but clearly originating 
in the Cyatheaceae. The extraordinary similarity of the vascular systems 
of P 'eranema, Diacalpe , and Nephrodinm Filix-mas has laid the end of 
that line in the Aspidieae ; the details in the description of the mature 
sporangia of Diacalpe and Peranema have confirmed the theory founded 
on the vascular system (Davie, T 2 , p. 264). And now in the account of 
the development of the sorus and sporangium of Peranema we have 
evidence which strengthens this view. 
Probably Peranema is not itself a link between the Cyatheaceae and 
the Polypodiaceae ; but features in its development, taken in conjunction 
with features of Woodsia and Diacalpe , suggest that the process of evo- 
lution has moved through types like the Cyatheaceae to types like those 
found in the Aspidieae group of Polypodiaceae. The account given above 
of the sorus, sporangium, and prothallus of Peranema permits us with 
confidence to assert that the phyletic line already traced by Professor 
Bower from the Gleicheniaceae through the Cyatheaceae (Bower, T 2 ) has 
proceeded through the Woodsieae-Woodsiinae group to the Aspidieae 
group of the Polypodiaceae. Further stages along this line probably 
moved from N ep hr odium (of whose sorus we now possess an interpreta- 
tion) through Aspidium and P oly stichum to certain types included in the 
comprehensive genus Polypodium . 
