ON THE STRUCTURE AND AFFINITIES OF METACLEPSYDROPSIS DUPLEX. 173 
transition between scalariform and reticulate thickening is often clearly seen (PI. I. fig. 2). 
The largest of these elements have rarely more than three rows of pores on each wall. 
The outer zone of the stem xylem is four to five cells deep and constitutes a ring of 
solid wood round the inner zone. The elements of this ring are much larger than those 
of the inner wood, but the tracheides towards the inner margin of the ring are smaller 
than the others asa rule. The average size is from 100 to 130», but near the bifur- 
cation, to be noted shortly, some of the elements are as much as 250» in diameter. 
Some of these large tracheides may be seen in Pl. I. figs. 4, 5, and 6. At the departure 
of a petiole-trace larger elements also occur (PI. I. fig. 8). Like those of the inner zone, 
the outer tracheides are long, pointed, and their walls reticulately thickened, but there 
are six to nine rows of pores on the walls. ‘he pores on the walls of the elements of both 
zones are about the same size. Groups of small tracheides may occasionally be seen in 
the outer xylem zone, but these are the decurrent protoxylem elements of a petiole-trace 
and do not belong to the stem itself. 
Outside the xylem there is generally a dark layer which in very thin section is seen 
to consist of cellular tissue. This tissue no doubt represents inner cortex and phloem, 
but it seems to have been composed of very delicate elements, which are too much 
erushed to warrant more than a passing notice. The outer cortex, on the other hand, 
is generally well preserved. It consists of a thick-walled parenchyma, but there are 
no sclerotic layers present. ‘his outer cortex is sometimes produced into ragged 
protuberances. 
To the stem xylem, petiole- and root-traces are attached, but they only occur at 
considerable distances from each other. In the material examined only three or four 
of such emissions have been noted, and in but one case was a root-trace found joining 
the stem. In the case of each of these appendages, however, there is sufficient evidence 
to show that the departure is protostelic, z.e. the outer xylem ring is never broken, 
thus exposing the inner zone on the surface of the cylinder. 
BRANCHING OF THE STEM. 
In several of the stems examined it was possible to cut long series of sections, but 
the changes were very slight, except in the neighbourhood of the emission of a petiole- 
or root-trace. In one specimen, however, just above the point of departure of a petiole- 
trace, I was fortunate enough to discover both a root-trace emission and a bifurcation 
of the stem, within a length of 14 inches. In another case, just above a petiole-trace 
departure the very small solid stele of Pl. IV. fig. 46 was discovered, and this is possibly 
an unequal dichotomy of the stem. 
Pl. L fig. 3 represents the first stage of such a bifurcation (the small trace rt. ér. 
represents a root-trace which arches over at this point, and the top of the arch is here 
shown), and the inner zone of mixed pith is distinctly elongated, as is the whole stem 
xylem. Roughly speaking, the stem xylem is an ellipse. Higher up the stem the 
