ON THE STRUCTURE AND AFFINITIES OF METACLEPSYDROPSIS DUPLEX. 171 
that it would have been referred to another genus, namely, Dinewron. Fig. 24 repre- 
sents another example where the trace is even more like Dinewron than that of fig. 23. 
The next two figures (25 and 26) are cut rather obliquely, and the traces are perhaps 
not really qnite so long as they appear, but the angle of section is very near 90°, so that 
any reduction in their size on account of the obliquity of section must be very slight. 
They are from different specimens, but represent stages which are practically identical 
and show the departure of a much reduced pinna-trace-bar (b). Fig. 26 is probably 
further up the petiole than fig. 25, and the bar (b) shows clearly that it will divide into 
two portions. A somewhat similar example, cut from the same specimen as fig. 26, is 
shown in Pl. III. fig. 27, though here the trace is rendered shorter than it really should 
be by a slight crushing. In Pl. III. fig. 28 the trace is still further reduced, but it is 
clearly identical with that of fig. 27, while that shown in fig. 29 is also similar. 
Pl. III. figs. 30 and 31 represent examples quite comparable with that shown in 
fig. 29. 
_ The next three figures represent sections taken from the same specimen, and the 
petiole joins the stem two sections below that shown in fig. 34. Fig. 32 is similar to 
fig. 31, except that both ends of the trace are closed instead of only one end, as in the 
former figure in which the left-hand end exhibits an island of parenchyma in the xylem, 
while the right hand end has a bay. Fig. 33 shows clearly the two almost circular 
islands, one at each end of the trace, and also the stem (st) which the petiole joins at 
a lower level. Near the junction with the stem the islands disappear and only the 
small protoxylem elements remain. These protoxylem groups occur in pairs at each 
end in fig. 33, but in fig. 34, where unfortunately the lower part of the trace is crushed, 
there only appears to be one group. ‘Two sections lower down in this series the petiole- 
trace joined the stem (figs. 33 and 34, st). The petiole-trace of M. duplex has thus 
been followed through all its stages into a stem which, while presenting certain 
peculiarities, is very simple in structure. 
A typical transverse section of this stem is shown in Pl. I. fig. 1 or Pl. IV. fig. 45. 
Two regions may at once be distinguished, an outer of large tracheides which are seen 
to be reticulately thickened, when viewed in longitudinal section, and a central zone 
composed of a mixture of tracheides and parenchyma. ‘The inner tracheides are long, 
pointed elements and have reticulate or scalariform thickenings on their walls, but 
they are smaller in diameter as a rule than those of the outer zone. The stem was 
long and dichotomously branched. 
On the whole it had the appearance of a rhizome, and evidence in favour of this will 
be brought forward immediately. 
Root-traces have been met with on one or two occasions, but only one example 
actually joined the stem xylem. One of these root-traces is shown in Pl. IV. fig. 44. 
| It is large for a root-trace and is diarch. 
So far merely the xylem tissue has been considered, but the cortex is also of some 
| importance ; for, while there is a sclerotic layer in the cortex of the mature petiole, there 
TRANS. ROY. SOC. EDIN., VOL. XLVIII, PART I. (NO. 8). 27 
