1089 
Relationships of the Gnetales. 
and of the typical ones ; there remains a more highly developed condition 
which is occasionally to be found, but which has not yet been described. 
It is illustrated in PI. XCV, Fig. 31, which is a photograph from E. distachya . 
The upper two perforations appear from their shape to have resulted each 
from the horizontal fusion of a pair of perforations. This inference is con- 
firmed by the condition of the pair of opposite perforations below, for the latter 
are separated by an extremely thin part of the wall (compare with Fig. 20). 
It is clear, therefore, that a fusion of perforations sometimes occurs, result- 
ing in the formation of still larger perforations. The resulting vessels bear 
a much closer resemblance to those of the Angiosperms, especially to the 
ones with scalariform end walls which are characteristic of the lower 
Dicotyledones. 
In spite of this resemblance, however, it must be admitted that the 
vessels of Ephedra differ essentially from those of the Angiosperms. The 
perforations of the Angiospermic vessels by no means correspond to modified 
bordered pits. It may be stated here, however, that the vessels of Gnetnm 
to a large extent clear up the difficulty. 
3. Wood Parenchyma. 
Wood parenchyma is very abundant in Ephedra , the number of its 
elements often nearly equalling the number of tracheides. It is sometimes 
scattered irregularly throughout the wood (PI. XCIV, Fig. 10), but is more 
often arranged in definite tangential bands. The banded arrangement is 
visible in PI. XCVI, Fig. 39, in which the parenchyma cells appear dark owing 
to their deeply staining protoplasmic contents. There are usually two rows 
of tracheides separating the successive bands of parenchyma cells, although 
often, especially in root wood, only a single row of tracheides intervenes. 
Sometimes, indeed, the bands of parenchyma cells themselves become two 
or three rows in thickness. 
The individual cells are identical with tracheides in length and shape 
(PI. XCV, Fig. 23). They are also lignified in the same way and to the same 
extent as tracheides. Moreover, in the vast majority of instances they have no 
cross-walls as do those of Conifers. Therefore, the only features which serve 
to distinguish them from tracheides are the character of the pitting and the 
possession of protoplasmic contents. When in dry material the protoplasm 
disappears, the distinction is often difficult to make. 
As for the other distinguishing feature, namely, the character of the 
pits, it is .usually simple. This is always true in the case of pits between 
adjacent parenchyma cells. In some cases, however, where the parenchyma 
cell adjoins a tracheide or vessel the pits show a small border on the 
parenchyma side as well as the usual large one on the tracheide side. The 
small border is shown in Fig. 32. The parenchyma cell adjoining the vessel 
