763 
Stele in Two Species of Dipteris. 
cells. Some sections later the central tissue has again increased and dark 
cells are formed, but afterwards they disappear again. The position in 
which these cells appear varies a good deal. Sometimes they are formed 
when a leaf-trace is about to depart, sometimes just before the xylem-gap is 
closed, but they may be absent from both these positions. Fig. 13, PI. LVIII, 
shows the stele of a young plant. The rhizome of this plant was 4 mm. 
in length and had formed eleven leaves. In this stage a leaf-trace has left 
the stele some little distance below, and the xylem is not yet quite 
closed. Two dark cells are seen in the centre surrounded by phloem. 
Fig. 14 is a section of the same rhizome after four more leaf-traces have 
gone off. No dark cells are present here. The xylem is much more 
developed on the ventral side. At the departure of the first leaf-traces 
there is no connexion whatever between the dark inner cells and the outer 
endodermis, but after some more leaf-traces have gone off, the dark inner 
cells come into connexion with the outer endodermis, when the leaf-trace 
separates from the stem-stele, and a real leaf-gap is formed for the first time. 
It may happen, however, that on the departure of the next leaf-trace there 
is again no connexion of the outer endodermis and the inner dark cells. 
The inner endodermis can now often be distinguished from the inner peri- 
cycle as a definite tissue, but it does not yet form a regular layer. The 
number of its cells constantly changes, and it may even disappear com¬ 
pletely. 
The presence of internal endodermal cells in the stele, before a real 
leaf-gap is formed and before any connexion has taken place between the 
outer endodermis and the inner tissue, shows that the inner endodermis is 
formed independently in the centre of the stele, and that the inner endo¬ 
dermal cells are not in any sense formed by ‘ intrusions ’ of the outer ones 
through the leaf-gaps. 
In a series from an older stem we find a true solenostele, but the xylem- 
ring is still quite obviously thicker on the ventral than on the dorsal side. 
Protoxylems are apparently absent, as is generally the case in the first- 
formed stems of Ferns. The pith consists in the first section only of one 
cell (Fig. 1, PI. LVII), and a few sections later no pith cell can be recog¬ 
nized. At the end of the series, however, the pith is quite a distinct tissue. 
Through the whole series the cells forming the pith and the internal endo¬ 
dermis are constantly decreasing and increasing in number; on the whole, 
of course, they increase. In the first appearance of the internal endodermis 
this fluctuation in development is already observed. It is seen in all the 
different tissues of the stele in the youngest stages, and here in the older 
stem it is again quite obvious, especially in the pith. The leaf-traces suc¬ 
ceed each other in this stem very rapidly, and are rather small in comparison 
with the stele. They mostly go off with two xylem-strands. 
In a still older stem the structure is much more like that of the mature 
