234 
Lady Isabel Browne. 
occurrence” in the Botryopterideae; but, as already indicated, 
endarchy may well have been exceptional in that order. He also 
contends that the advantage of an early development of peripheral 
tracheides for the attachment of roots might become a factor in 
evolution, but that it is very hard to conceive of endarchy as the 
derivative condition. It has, however, already been recognized that 
centrifugal wood has frequently in the phylogeny supplanted 
centripetal xylem. 
Mr. Boodle’s view that the sub-collateral type of stele arose by 
reduction from the two-banded type of T. reniforme seems well- 
founded. He was able to show that in some species the larger 
examples are two-banded, and the smaller ones sub-collateral. 
Thus the available data tend to disprove the theory that the simple 
collateral types were primitive. Moreover it is clear that the truly 
collateral type of stele, i.e., that in which the phloem is confined to 
one side of the xylem, arises by a continuation of the process of 
reduction by means of which the sub-collateral type was derived 
from the two-banded one. This is borne out by the fact that many 
of these species with collateral steles develop very little xylem (3). 
Mr. Boodle considers it very probable that the Hymenophyllaceae 
were derived from non-filmy ferns. This is strongly supported by 
the presence of pseudo-veins, devoid of vascular tissue, and by the 
facts quoted, showing that there has been a considerable reduction 
in the steles of many species. It is also borne out by the fact that 
the filmy habit appears to have arisen independently of the Hymeno- 
phyllaceae in the Osmundacem. In spite of so many indications of 
reduction Professor Hallier regards the one-layered condition of 
the leaves of most Hymenophyllaceae as primitive, not only for the 
order, but also for the phylum (18). 
There has been much discussion as to whether the flat thalloid 
prothallus found in Hymenophyllum and some species of Trichomanes, 
or the filamentous and Alga-like one of other species of Trichomanes , 
is primitive for the order. Prantl and Goebel considered the fila- 
mentons form as primitive, and Sadebeck has recently re-stated this 
view, asserting that although there is in the order a beginning of 
the formation of a cell-surface, “ . . . yet we see from the develop¬ 
ment as a whole that the filamentous form of prothallus is the 
original one for Trichomanes ” (22). Professor Bower and Dr. 
Campbell, on the other hand, attribute to the filamentous prothallus 
a secondary origin from the thalloid type, and the latter states that 
its form is probably largely due to environment (12), (7). This view 
