PHYLETIC RELATIONS 655 
of sorus: the one with marginal sori, the other with superficial. They 
are both characterised by having their spore-output restricted to the typical 
number of 64 as the upper limit. It may be a question whether these 
two were truly distinct in phyletic origin, but they appear to have pro- 
gressed independently. The one comprises the Dennstaedtia-Davailia series, 
leading onwards to various families of the Mixtae: the other is the Cyathe- 
aceous series. Besides these the genus Dzfzeris seems to have taken a 
line of its own towards the mixed sorus, and may perhaps be not alone 
in having done so. 
The most complete phyletic story has been made out for the Denn- 
staedtia-Davallia series, to which Dicksonéa itself may probably be a collateral 
line, with unusual elaboration of the structure of the stock. The scheme 
suggests a somewhat independent origin for this series, together with such 
genera as Loxsoma, Thyrsopteris, and Dicksonia, from a source probably 
between the Botryopterideae and Schizaeaceae, though not directly from 
either of them. They all show a structure of the axis advanced to a 
solenostelic or even a dictyostelic state, and a basipetal sorus, derived pre- 
sumably as an ameridment on the simple type: Zoxsoma and Thyrsopteris 
have a peculiarly archaic structure of the sporangium. The further pro- 
gression, which harmonises reasonably with the palaeontological record, 
has. been traced both anatomically and sorally, and leads with advance 
in both respects, through the Dennstaedtia-Microlepia-Davallia sequence 
from Ferns with a solenostelic structure and basipetal sorus to those 
with a dictyostelic stock and mixed sorus. Several side-branches, or it 
may be concurrent lines, also exist, such as Lzndsaya, Pteris, and Adiantum, 
thus giving rise to the bulk of those Ferns which Prantl grouped together 
as the Pterideae. 
But in several branches from this line a transition is found from the 
marginal to the superficial position of the sorus. One line is direct from 
Dennstaedtia through Hypolepis, a genus with mixed sorus, to some forms 
with superficial sorus reckoned as Polypodium: another line may have been 
as illustrated by Departa, where the mixed sori are sometimes marginal, 
sometimes superficial. Another, and a more decisive line is through such 
forms as Cystepteris, and certain sections of Davadllia such as § Leucostegia, 
in which there can be little doubt that the mixed sorus has been shifted 
from the margin to the surface. And so by a number of phyletic sequences, 
by no means exhausted by the examples quoted, it seems probable 
that there has been progression to that prevalent and final type of the 
Mixtae which has an intra-marginal sorus on the lower surface of the 
leaf. The Ferns thus derived are characterised by their advanced 
dictyostely. : 
But there is some reason to think that a similar result has been reached 
also by a more direct route. The series of the Cyatheaceae is linked by 
Alsophila, and especially by the solenostelic 4. prucnata, with the Glei- 
cheniaceae: it is a very slight transition, already indicated by G. dichotoma, 
