294 
PROFESSOR W. C. WILLIAMSON ON THE ORGANIZATION 
rootlet so far as is indicated by its size. Excepting in the magnitude of its liber-cells, 
the resemblance to the corresponding organs in the Selaginellece is complete. Seeing 
that this peculiar structure only exists in the recent Lycopods and Opliioglossece, and 
that no other resemblance exists between the fossil Lepidodendra and Sigillarice and 
the Opliioglossece, we must fall back upon the Lycopods as the plants with which this 
form of rootlet indicates true affinities. On forwarding characteristic specimens of 
these rootlets to M. Van Tieghem he at once replied, expressing his conviction that 
they belonged to a Lycopodiaceous plant.* 
It is now r fully established that Stigmaria ficoides, to which these rootlets belong, is 
the root of Sigillaria, as well as of Lepiclodendron. The structure of these rootlets 
therefore adds an additional link to the long chain of evidence, indicating the close 
affinity of the Sigillarice with the Lycopods. 
But there is yet another feature in these exogenous Lepidodendroid stems which 
may prove to have some bearing upon the problem of evolution : this is, the presence 
of numerous barred cells in the medullary rays, to which fact T have already made a 
slight reference. In my memoir. Part II. (Plate 27, fig. 23, a , p. 236), I called atten¬ 
tion to the existence of such cells in the triangular spaces that appear to project out¬ 
wards from the medullary sheath, and are intercalated between the inner extremities 
of the vascular wedges of the exogenous zone ; and in the same memoir (p. 238) I 
further pointed out that such cells entered into the composition of the medullary rays. 
This is not only the case with the Arran plant described in my last memoir, but I now 
discover them to be abundant in the various exogenous states of Lepidodendron 
selaginoides, and to be especially conspicuous in the peripheral layers of the more 
matured branches. Fig. 21, h, represents peripheral portions of two medullary rays of 
fig. 7, enclosing between them a single wedge of the barred vessels. Fig. 22 represents 
a portion of another, but similar, section, taken from the point where the exogenous 
zone, h, joins the medullary sheath, a. A number of small vessels, x, x, appear in all 
sections similar to the latter one, respecting which it is difficult to determine whether 
they are to be regarded as belonging to the medullary cylinder or to the exogenous 
zone. Where the medullary rays originate amongst these small vessels, oblong barred 
cells enter conspicuously into their composition, apparently associated with cells that 
are not barred. The increase in the number of these barred cells as we follow the 
course of the ray centrifugally is accompanied by a corresponding increase in its 
diameter as seen in such transverse sections as fig. 7. 
CoRDA found similar cells in the medullary rays of his Diploxylon cycadeoideum, 
as pointed out in my memoir, Part II. ; but he did not distinguish these rays from 
foliar bundles, which he believed them all to be. Tangential sections, however, show 
* “ Par la structure du cylindre central, et par la division dichotomique qu’on y observe en plusieurs 
endroits, vos racines appartiennent bien certainement a une Lycopodiacee de lafamille des Selaginellees.” 
In liter a, Oct. 30, 1879. 
