235 
Anatomy of the Ophioglossaceae . I. 
however, consistent with and support it, especially if the structure be 
considered from above down, and not, as in the descriptions given, from 
below up, Down to a level which probably marks the position of the 
original bud, the structure of the branch is normal. Below this we come 
to the vascular connexions to the pre-existing tissues of the main shoot. 
These must be regarded as developed, under the influence of the developing 
bud, on the tissues occupying the axil of the leaf-trace. Further study of 
the anatomy of such structures may throw much light on the immediate 
causes of the organization of the vascular system. The remarkable fact 
comes out from the study of these few branches that these secondarily 
established vascular connexions are such as to allow of comparison with 
other lateral branches, the vascular supply to which is part of the primary 
organization of the plant. 
While the relative position of the branch remains the same, its vascular 
supply may appear to come almost wholly from the subtending leaf-trace, 
or may in considerable part be derived from, or connected with, the tissues 
of the stem stele. The former case is best shown in the first branch, less 
clearly in the second, while connexion with the stem stele is marked in the 
third and fourth specimens. This variation in Botrychinm Lunaria is of 
considerable interest when it is borne in mind that the branch in Helmin - 
thostachys is connected with the stele of the stem at the anterior end of the 
leaf-gap[and not with the subtending leaf-trace. 
On considering all the facts, I am inclined to regard the vascular 
relations of the branch in the first specimen as most nearly representing 
what would be found if the development of the branch were immediate 
and primary. The endodermal relations show that this bud must have 
sprung into activity in a region so close to the apex of the main shoot that 
the endodermis was not yet differentiated. The preparations for the 
vascular supply of this branch at once suggest comparison with the more 
or less complete development of adaxial xylem in normal leaf-traces. It 
seems possible that the structure in the second and third examples can be 
brought into line with this ; at any rate, whatever be the significance of the 
xylem connecting the branch and stem steles in these three cases, the main 
relation of the branch is to the adaxial xylem of the subtending leaf-trace. 
The fourth example, however, seems to show that the vascular connexions 
may be more irregular and not always reducible to one plan. 
While all the details cannot be reconciled it seems to me that we are 
justified in looking on Botrychinm Lunaria as having ancestrally or poten- 
tially a branched shoot, the branches springing from the leaf-axils. It seems 
justifiable further to compare the origin of the vascular supply to the branch 
with what is found in the branching of the Hymenophyllaceae and of some 
of the Zygopterideae. That variants should occur in Botrychium is not 
surprising ; indeed, it is remarkable that the subsequent establishment of 
