12 
PEOEESSQE W. 0. WILLIAMSON ON THE OEGANIZATION 
they belong, as might be expected, to the barred type. They are very distinctly pre- 
served in a transverse section of one rootlet, in which the bundle consists of about a dozen 
vessels, of which the largest have a diameter of -0025, whilst the smallest are not more 
than -0008. This bundle is represented in fig. 22. The epidermal hairs (figs. 20 & 21, k) 
are extremely numerous and well defined; they are cylindrical, and composed of a 
single linear series of cells with rectangular transverse septa, which are often indistinctly 
seen. Their diameter averages -0017. The length of the cells is variable, but those 
which I have measured are usually about ‘015. Fig. 19, b" shows that these hairs are 
but extensions of the outermost interradicular parenchyma (b'). 
In the absence of all trace of the more central and vascular portions of the main axis 
to which these adventitious rootlets have belonged, it is impossible to determine to what 
class of Ferns they belong*. In the form of the vascular bundle of the rootlet it 
* As stated above, when my memoir was written I bad found no trace of the vascular axis of the plant here 
described ; but the day before this memoir was laid before the Eoyal Society Mr. Carruthers kindly showed me 
some sections of the plant from his cabinet, amongst which was one of which fig. 22* is a representation, of the 
size of nature. In the upper left-hand corner of this section there is preserved a small portion of a central stem. 
This part of the section is further enlarged ten diameters in fig. 22**, in which cl d is a narrow lamina of vessels 
which appears in the transverse section as a semicircular band. These vessels are not arranged in any regular 
radiating order, and exhibit no trace whatever of having received any increase to their number from exogenous 
growths. The character of the vessels does not appear in the transverse section, and unfortunately no longitu- 
dinal one appears to have been made of the specimen. The vessels vary in size ; hut the greater number of 
them have a mean diameter of from -0033 to '005, some few being larger and others smaller than these dimen- 
sions. This vascular belt is interrupted in the middle ; but whether this break is an accidental rupture, or 
whether the two halves represent parts of separate crescentic masses, is not certain. This vascular lamina has 
evidently been imbedded in a mass of very regular parenchyma, an inner portion appearing at a , whilst frag- 
mentary portions remain of a peripheral mass which has occupied the interval between the vascular layer and 
the periphery of the stem from which the numerous rootlets (i) have sprung. The inner parenchyma (a) and 
the inner portion of the outer parenchyma correspond very closely, and consist chiefly of cells that have a dia- 
meter of from -0025 to '005. But the more peripheral portion of the outer tissue consists of a dense mass of 
very regular, small, thick-walled cells, with a diameter of from ’001 to '0006, approximating closely in their 
appearance in the transverse section to the fibres constituting the outer layer of each rootlet. A curious feature 
of these parenchymatous structures is the presence in them of numerous longitudinal canals (c, c), which appear 
to be identical with the gum-canals of the Myelopteris described in the earlier part of this memoir. 
Whether the stem originally contained a single circle of these vascular laminae, or whether there were suc- 
cessive layers concentrically arranged, as in many species of Psaronius figured by Corda, is not determinable 
from Mr. Carruthers’s sections. The great probability, however, is that the latter condition has existed. It 
will be observed that this, the only stem of a tree fern from the British Carboniferous deposits of which the 
internal organization has hitherto been described, exhibits the same absence of all exogenous growths as that 
to which I called attention in some of my previous memoirs as characterizing the continental Psaronites. So 
far as my present experience has extended, all the fossil ferns had, like their recent allies, closed vascular bundles ; 
whilst the Calamites and Lycopods, including in the latter group the Aster ophyllites, had open ones. If Lygi- 
nodendron and Heterangium ultimately prove to be stems or rhizomes of Eerns, of course this generalization 
will have to he discarded ; but I have as yet obtained no such proof. 
Mr. Carruthers’s cabinet contains a beautiful section of Myelopteris, in which the outer vessels of each vas- 
cular bundle are small, but with unusually thick walls, as seen at d! in fig. 7*, which represents one of these 
bundles, with a small portion of the surrounding parenchyma, much enlarged. 
