549 
Origin of Medullation in the Ophioglossaceae. 
large size and massive structure. It presents at once to the sporeling an 
ample supply of nutritive material. The sporeling in this species is, as 
might be expected, relatively complex in its early structure. The pro- 
tostelic state is not represented in it, but it is medullated from the first. 
In B. Lwiaria the prothallus is described by Bruchmann 1 as being i mm. 
to 2 mm. long, and i mm. wide. It is very much smaller than that of 
B. virginianum , and will offer a correspondingly smaller nutritive supply 
to the sporeling. We should expect from this that its sporeling would be 
structurally less complex. It is seen to be commonly though not always 
protostelic. The medullated state is certainly not the primitive condition. 
It may safely be concluded that the young plant of B. virginianum , like 
the offspring of the rich, is saved from early struggles. It omits the 
elementary structural stages, and springs at once into a condition of 
structural advance. If then we desire to see in their fullness the early steps 
of ontogeny, it is not to the plethoric sporelings we should go, but to those 
that are less well nourished, or it may be even starved. In them we may 
expect to see the steps of preliminary structure most gradually taken, and 
in extreme cases drawn out almost like the joints of a telescope. This was 
in fact found to be the case in the sporelings of B. Lunaria from the high 
ledges of the Breadalbane Hills. It is believed that in such depauperate 
plants the successive steps of the ontogeny serve as a safe guide in studying 
the origin of the vascular structure seen in the mature plant. 
In the first place it is to be noted that in the early stages of these 
plants, though the external endodermis is well represented, an internal 
endodermis is absent. Sometimes it does not appear at all: in other cases 
it appears relatively late, in more or less direct relation to the intrusion of 
foliar pockets. But these are entirely absent from the first stages of the 
young plant. The medullation precedes them. Plainly, therefore, the 
pockets cannot cause the medullation. The converse is the more probable 
line of causality, viz. that the prior existence of soft internal tissue will 
favour the formation of intrusive pockets. From the study of these plants 
it is seen that the pith is in the first instance intrastelar, but it may sub¬ 
sequently be added to by the intrusion of foliar pockets. Sometimes the 
limits of these two components of the ultimate pith may be maintained, as 
shown by PI. XLV, Fig. 9, and indicated by Poirault (Figs. 11, 12, p. 169, 
&c., loc. cit.). But, as is well known from the time of Van Tieghem’s 
description of it, the endodermal limits become less definite in fully 
matured plants of Ophioglossum and Botrychium , while in Helminthostachys 
an internal endodermis is only known in strong and mature rhizomes. 
It appears that a like condition has ruled in the Osmundaceae, but it 
is attained only in a more advanced state of the young plant. Here the 
sporeling is from the first less bulky, perhaps in relation to its assimilating 
1 Flora, 1906, p. 205. 
