550 Bower.—On the Primary Xylem , and the 
prothallus, so different as a nursing mother from the bulky storage pro- 
thallus of the Ophioglossaceae. And the first ontogenetic stages are 
accordingly shown in more extended sequence. Faull points out the 
entire absence of foliar pockets in the young plants of Osmunda cinnamomea . 
Similarly Kidston and Gwynne-Vaughan have shown their absence from 
the earlier related fossils, such as Zalesskya and Thamnopteris , 1 But later 
in the development of the seedling of O. cinnamomea narrow foliar pockets 
make their appearance. And similarly in the later fossil forms, such as 
Osmundites skidigatensis and Kolbei , foliar pockets are found. Finally, in 
older plants of O. cinnamomea a continuity is established between the 
internal endodermis and phloem and the external, through the ramular, 
and less completely through the foliar gaps. 2 But this connexion only 
occurs in the mature plant. The ontogeny thus indicates that it is a late 
and derivative state. A like state is found in the relatively late fossil 
O. skidigatensis. The parallelism which thus rules between the ontogeny 
of the living forms of Osmundaceae and the story as based upon the study 
of the sequence of the fossils gives mutual strength and conviction to the 
evidence derived from each. From both sources the facts indicate that an 
up-grade development has occurred, involving a formation of intrastelar 
pith, and a progressive formation of leaf-pockets. But the latter seem never 
to have attained to a large size in the Osmundaceae, or to have contributed 
in the same measure as in the Ophioglossaceae to the sum of the central 
parenchymatous mass. This is the structural expression of the fact that 
in the Osmundaceae the leaves are numerous and individually smaller 
in proportion to the axis, while in the Ophioglossaceae, as a rule, only 
a single leaf is matured in each season, but it is of relatively large size. 
Summary. 
1. The young plant in all the genera of Ophioglossaceae is variable 
in its state of medullation. In some cases there is at first a solid xylem-core ; 
in others the pith is present from the first. 
2. This difference in medullation may be found between specimens 
of the same species, and there is reason to believe that it is determined by 
nutrition, the best nourished plants having a pith from the first. 
3. Where the axis in Ophioglossum and Botrychinm is at first proto- 
stelic, the pith is initiated below the departure of the first leaf-trace, and 
is referable in origin partly to intraxylic parenchyma, partly to the 
conjunctive parenchyma of the xylem-sheath ; the two tissues being put 
into communication at the first leaf-gap. Meanwhile, in Botrychiam the 
endodermis forms an uninterrupted barrier, shutting off the cortex; thus 
the medullation is wholly intrastelar. 
Fossil Osmundaceae, Parts II and III. 
2 Faull, Bot. Gaz., vol. xxxii, p. 417. 
