Anatomy of the Osmundaceae. 533 
The other Interpretation of the nodal pockets advanced by Faull 1 is 
that they do not indicate a primitively gapless departure of the leaf-trace 
(cladosiphony), but a ‘ cladosiphony secondarily produced and that they 
afford strong proofs of an evolution that tends towards cladosiphony. He 
holds that they owe their existence to an evolutionary tendency on the 
part of the xylem to increase in the direction of the pith. In fact, they are 
portions of the central ground tissue that have been closed in as a result ot 
a centripetal proliferation of the xylem. As instances of this tendency to 
centripetal proliferation, Faull mentions cases in Osmnnda cinnamomea in 
which he found xylem-strands of considerable size on the inside of the 
xylem-ring and separated from it by a varying amount of parenchyma ; 
also cases in which isolated tracheides, each surrounded by a ring of endo- 
dermal cells, occurred in the central ground tissue inside the internal endo- 
dermis. The latter he regards as having been ‘ pinched off’ from the stele. 
If this is really the case they should be connected at some point with the 
main xylem-ring, and they would then have some analogy to the internal 
accessory strands in Dicksonia adiantoides and D. rubiginosa , and to me 
they would suggest an increase rather than a reduction of the vascular 
tissue in the stele. I consider all these cases to be very interesting, but 
I regard them as indicating a readiness on the part of the cells of a true 
pith to revert to the type of element from which they were derived, in 
fact, as partial reversions to a mixed pith. The existence of stems with 
a parenchymatous pith containing scattered tracheae, isolated or in groups, 
such as occur in the Lepidodendreae, Zygopterideae, and Osmundaceae, 
seems to me to be a very serious obstacle in the way of the acceptance of 
Professor Jeffrey’s statement that all piths are extra-stelar . 2 It is possible 
to account for them on the lines of this theory by using Faull’s conception 
of a centripetal proliferation of the xylem into the enclosed extra-stelar 
ground tissue. But this is putting a heavy strain upon the idea and does 
not appear to me at all satisfactory. In particular, as regards the Osmun¬ 
daceae, I find it difficult to reconcile a proliferation of xylem with the theory 
advocated by both Jeffrey and Faull, that the vascular system of this order 
has undergone a simplification of structure owing to reduction in size. 
The Zygopterideae, again, present a further obstacle to those who deny 
the existence of a true intrastelar pith. For, so far as I am aware, no 
Zygopterid stem has yet been found with either a leaf-gap or a branch-gap 
in the stele, whereby the extra-stelar tissues could get in. 
The presence of short relatively thin-walled tracheae occupying the 
position of a pith in Thamnopteris , Zalessky a, and Diplolabis Rbmeri , 3 fits 
1 1. c., p. 524. 
2 Jeffrey : The Pteropsida. Botanical Gazette, vol. L, 1911, p. 401. 
8 Gordon : On the structure and affinities of Diplolabis Rome 7 'i (Solms). Trans. Roy. Soc. 
Edin , vol. xlvii, 1911, p. 711. 
