222 Lang . — Studies in the Morphology and 
commenced to reappear and from now on persists and enlarges. Not till 
the level reached in Phot. 1 5, where the small leaf-trace is dying out in the 
cortex, does the pith communicate with the leaf-gap parenchyma. The 
relations in this case are essentially similar to those described by Gwynne- 
Vaughan in Osmunda. The plant from which Phots. 12-15 were taken 
was further of interest in that after the stele had expanded to a medullated 
stele like that shown in Phot. 1 6, it again contracted and had scattered 
tracheides in the small pith. 
It was pointed out above that occasionally tracheides may be developed 
in the pith of normal plants. A more extreme case of the same kind was 
met with in a small injured stem, which bore a branch. This will be 
described later on, and it is sufficient to refer to the reconstruction of this 
specimen (Text-fig. 11), to the series of transverse sections (Text-fig. 10), 
and to Phot. 31. These show that the development of tracheides internal to 
the position of the protoxylem had led to this region of the stem possessing 
a veritable mixed pith of tracheides and parenchyma. 
The analysis of the xylem of Botrychium Lunaria in different regions 
of the rhizome thus leads to the recognition of secondary xylem, centri- 
fugal (peripheral or outer) primary xylem, which is constantly present, 
and centripetal (central or inner) primary xylem, which is more or less 
clearly indicated. This brings the construction of the primary xylem of 
Botrychium into line with that of Helminthostachys , where centrifugal and 
centripetal primary xylem are well marked. That the centripetal xylem of 
B. Lunaria when present usually disappears opposite a departing leaf-trace 
and takes no part in its formation only strengthens the comparison with 
the centripetal primary xylem of Helminthostachys. 
The leaf-trace of Botrychium Lunaria is usually and correctly described 
as departing from the stele of the stem as a single endarch strand of xylem, 
accompanied on its abaxial side by conjunctive parenchyma, phloem, peri- 
cycle, and endodermis in the same succession as these tissues are found in 
the stele of the stem (Phots. 3, 16, 17). On passing from the stem into the 
leaf-base the single trace divides into two strands (Phot. 21), which become 
inclined so as to face one another. The leaf-trace, however, exhibits con- 
siderable variety in its construction, and some of the main features observed 
will now be referred to. 
When the stele of the stem has marked secondary thickening, the 
leaf-trace on its separation shows, outside the arc of primary xylem, 
secondary tracheides arranged in radial rows and separated by medullary 
rays. This secondary xylem diminishes in amount and disappears as 
the leaf-trace nears its departure from the stem (cf. Phots. 3, 20). From 
this point onwards the xylem of the leaf-trace is altogether primary, 
and is continuous, when traced downwards, with the primary xylem of 
the stem stele. When the trace departs from a stele without secondary 
