Bower. — On Mednllation in the Pteridophyta. 565 
living genera of Osmundaceae. But such pockets are absent from the more 
ancient forms. This suggests that there has been a progressive formation 
of such pockets. On the other hand, in the earlier types there is merely 
a differentiation of the massive xylem-core into a centrally-lying region 
composed of short, thinner-walled elements, and a peripheral ring of firmer 
wood ( Thamnopteris ). In later forms, such as Osmundites Kolbei , the centre 
was occupied by a ‘ mixed pith ’, containing tracheides, while the remaining 
xylem consisted of separate strands. 1 This leads naturally to the condition 
of the modern types, with their central pith of parenchyma, but occasionally 
containing tracheides, as shown by Faull. 2 The intrastelar pith is fore¬ 
shadowed in the earlier fossil types, and the stratigraphical sequence 
illustrates its origin along lines in accordance with the theory. Thus the 
actual structure of the Osmundaceae, ancient and modern, is such as to 
accord with what might have been anticipated, supposing the suggested 
relation of habit to internal structure to be valid. 
Similarly, in the case of the Ophioglossaceae, probably their stock 
was primitively upright, and they are not only megaphyllous, but they 
develop commonly one leaf only in each season, and that of a large size 
(monophyllous). Supposing the suggested relation to hold, they should 
have an intrastelar pith in accordance with their upright stock, together 
with intrusive pockets in accordance with their megaphyllous character. 
Since in Ophioglossum the endodermis is less perfectly developed than in 
Botrychinm , it is from the latter that the best evidence may be derived. 
It has been seen in the preceding paper (p. 543) that a pith, which is at 
first entirely intrastelar, arises partly from the intraxylic parenchyma, partly 
from intrusion of the conjunctive parenchyma. In weak plants this con¬ 
dition may be continued for some distance, and appears to correspond 
in origin and in structure with that demonstrated by Faull in young plants 
of Osmunda cinnamomea . But the intrastelar pith thus formed is liable to 
be encroached upon by intrusive foliar pockets, which become successively 
larger with the enlargement of the leaves. The endodermis, yielding 
1 Professor Jeffrey (Bot. Gaz., Dec., 1910, p. 476, &c.) meets the evidence adduced by Kidston 
and G Wynne-Vaughan from Osmundites Kolbei with the remark that ‘ the unprejudiced anatomist 
would scarcely admit the accuracy of their statements on the evidence they submit \ Doubtless the 
authors may have something further to say on this. Meanwhile, the ‘unprejudiced anatomist’, 
of whom we have heard so much in these discussions that we begin to doubt his existence, will draw 
his own balance between the criticisms of a writer at a distance from the specimens, who has a very rigid 
statement to defend, and the carefully considered conclusions of two experienced observers, whose 
study of the fossils themselves has been the basis of their statement. Moreover, he will take fully 
into account the coherent and consecutive features of their evidence, of which the facts for Osmundites 
Kolbei are only a fragment. Faull and Sinnott also take the same line of defence of their theoretical 
position as Professor Jeffrey by discounting the evidence from the fossils (which they have never seen) 
on ground of their imperfect preservation. Those who are accustomed to the interpretation of fossil 
structure will probably be able to assess at their proper value these criticisms in absentid . See 
Faull, 1 . c., p. 530, and Sinnott, 1 . c., p. 107, &c. 
2 1- c., p. 5?5- 
