369 
between this group and the WMarattiaceae. The apical cell in 
B. neglectum is very suggestive of that found in Danaea, and the 
reduplication of the leaf trace from the Ophioglossaceous to the 
Marattiacoous condition 1s another significant point. 
4. Botrychium can be divided into two groups : Phyllotrichium, 
in which the leaf dominates ; Eubotrychium (a better name would 
be ‘Radicotrichium”) in which the root is the dominating 
structure. Osmundopteris seems to hold an intermediate position. 
The general relationship of Botrychium, apart from the points in 
common with Ophioglossum and Helmintostachys, points in the 
first place to the Marattiaceae. It is evident that these two 
groups are closely interrelated. Other evidence points to the 
Osmundaceae. These three groups of ferns have many charac- 
teristics in common. Significant in this connection is the fact 
that the endophytes as described by Campbell for the Ophio- 
glossaceae, by West for the Marattiaceae, and by Cribbs for 
the Osmundaceous Todea, are probably identical. The presence 
of Stygeosporium marattiacearum therefore seems to be esta- 
blished for these three groups. 
We will refrain from speculations on the phylogenetic rela- 
tionship of the different forms treated in this paper. As soon 
as we assert a certain form to be primitive, we follow our 
personal taste more than scientific deductions. So - called 
primitive characters are widespread in Botrychium, but they 
seem to be distributed more or less equally throughout the 
genus. 
