6 
PROFESSOR W. C. WILLIAMSON ON THE ORGANIZATION 
The same appearance is observable in fig. 14, where, at e, we find a similar sheath. 
This latter bundle belongs to the specimen fig. 1. In some few instances I find a few 
prosenchymatous fibres entering into the composition of the sheath ; but this is far from 
being a constant feature. 
I have already described the numerous isolated gum-canals which enter into the 
structure of these petioles ; but besides these we have numerous others associated with 
the vascular bundles. In the section (fig. 1) it will be seen that every one of the vas- 
cular bundles has connected with it a large round or semilunar orifice. In the enlarged 
fig. 14 one of these orifices (d) is clearly seen to be enclosed within the cellular bundle 
sheath e. At first I was convinced that each of these was a large gum-canal which 
formed part of the bundle. Fig. 9 represents a small portion of one of several sections 
which I made of one specimen, in which very few bundles exhibit any trace of similar 
canals closely associated with the bundles, though such canals exist independently 
(fig. 9, c ) in large numbers and of conspicuous size. On dissecting Captain Aitken’s 
specimen, I found that many of its bundles were as devoid of open spaces as fig. 9. 
Others had semilunar lacunae like those of fig. 1, whilst others were so exceedingly large 
and irregular as to form a cavity entirely surrounding the vascular bundle. One of these 
irregular orifices is seen in fig. 7, d, d ; and on comparing it with the small gum-canals 
( c , c) in the same figure it becomes obvious that they are different structures. These 
and other similar facts led me to distrust my first conclusions, and made it probable that 
these open passages were rather spaces caused by the detachment of the vascular bundles 
from the surrounding cellular tissue, such as we frequently observe in sections of the 
recent Lycopods. Further studies of these sections led me to conclude that the expla- 
nation may be found in a combination of these hypotheses. Thus in fig. 8, c we have 
an unmistakable gum-canal associated with the large bundle represented there. In 
fig. 5, c similar combinations exist. In fig. 10, on the other hand, which is a longitudinal 
section of one of the bundles belonging to the same specimen as fig. 8, we have no gum- 
canal. Yet, as I have already shown, we cannot for a moment believe fig. 7, d to repre- 
sent, in any sense, the small canals ( c , c) of the same figure. I conclude, therefore, that 
such canals are wholly absent from some bundles as in fig. 9 ; that they are distinctly 
present, but of small size, in such bundles as fig. 8, c ; that they exist in similar position, 
but of much larger dimensions, in such cases as figs. 1, 2-6, & 14 ; and that in examples 
like fig. 7, such a canal, devoid of any true walls (being, in fact, a mere intercellular space), 
has constituted a weak point, which has been converted into a larger irregular cavity 
by the shrinkage of the neighbouring cellular tissues*. 
There yet remains to be considered the cortical layer of this curious plant. Here 
again we meet with variations in different specimens. In very young rachides, like 
* Since writing the above I have dried thin sections of the recent petioles of Angiopteris erecta between two 
plates of glass, and found that lacunae were formed contiguous to each vascular bundle exactly corresponding 
with those seen in my fossils. This observation puts the origin of these large lacunae in mechanical shrinking 
of the parenchyma surrounding the vascular bundles beyond the reach of doubt. — June 1st, 1876. 
