170 W. T. GORDON 
In following out the changes in the petiole-trace much greater difficulty has been 
met with, and the stages figured are not from long series. This was due to the fact that 
the material consisted of short lengths of petioles, no specimen exceeding 6 inches. In 
that distance the changes were very slight, so that long—very long—series would _pro- 
bably be necessary to show the gradual change from one of the stages figured, to the 
next. In spite of this discontinuity, the position I take up will be readily accepted, for 
there is sufficient evidence to prove the general problem, namely, that there is a gradual 
9% 
ce 
disappearance of the “waist” * or constriction in the middle of the hourglass-shaped 
petiole-trace, and that this reduction takes place as we descend in the petiole. The 
last part of the problem is much more difficult to prove than the first, but, concurrently 
with the disappearance of the “ waist” of the petiole-trace, there is a marked reduction 
in the size of the whole trace and in the pinna-traces which join it. We know that the 
petiole-trace in other ferns gradually becomes smaller as we descend towards the stem, 
and that there are reduced pinna-traces towards the base of the petiole. This is what 
one would naturally expect. It is not surprising, therefore, to find a similar condition in 
Metaclepsydropsis. But apart from the xylem tissue, we find that the outer cortex 
loses its sclerotic outer layer as we follow this series. Now, from a study of the stem 
cortex, which has no sclerotic elements, we should naturally expect that the free petiole 
near its base would have no sclerenchyma in the outer cortex. From this evidence also 
we must conclude that the series, as figured, shows the various stages in a descending 
order. In other words, the proof of the problem lies in the following observations :— 
The size and shape of the petiole-trace, the development of the pinna-trace-bar and— 
pinna-traces, and the sclerotic outer cortex are all gradually reduced as we proceed 
through the figures shown from PI. IT. fig. 17 to Pl. III. fig. 34. 
It might be said that this was not conclusive, since all the figures do not represent 
different levels in one and the same specimen, and that there are perhaps several — 
different species mixed up in the series. Such an objection can easily be met, for each 
stage figured is intermediate between its two neighbouring figures; there is no sudden 
jump at any point from one type of trace to a totally different one. In fact, there is a 
continuous variation in one direction, and this is the only variation that is shown by the 
specimens. 
Pl. II. fig. 17 represents a petiole-trace perfectly typical of the species Metaclepsy- 
dropsis duplex; the waist is very pronounced, and the pinna-trace-bar is also well 
marked. The specimen shown in fig. 18 is also perfectly typical though rather smaller, 
but, apart from a few details in the phase of the pinna-trace departure, the two figures 
represent specimens which are specifically identical. In fig. 19, however, the waist is 
much less marked than in figs. 17 and 18, while fig. 20 exhibits a trace with hardly 
any constriction in the middle. There is still a slightly smuous appearance in fig. 21, 
but in figs. 22 and 23 this has entirely disappeared. Indeed, had the petiole whose — 
trace is shown in the last-mentioned figure been discovered separately, there is no doubt 
* The word “ waist” is used here as it gives the idea of a constriction in the middle of the trace, 
