226 
Weiss. — A Stigmaria with Centripetal Wood . 
Stigmaria Jicoides. In their origin these rootlet bundles will be seen to 
agree with those figured by Renault (’ 93 ) in a Stigmaria from the Champs 
des Borgis, near Autun (see Renault, PL XL, Figs. 8 and 9). This Stigmaria, 
like the Stigmaria Brardii , has centripetal wood. The rootlets in coming 
off did not break the ring of primary wood, but carried with them mainly 
the protoxylem elements. 
Towards the outside the bundles become more oblique, as can be seen 
from the one included in Fig. 2, PI. XV of the present communication. This 
oblique course of the rootlet bundles obviously prevents the cylinder from 
being split up in the manner usual in Stigmaria jicoides. Renault’s 
Stigmaria Brardii seems to agree with our Stigmaria in this particular too. 
In tangential longitudinal section the rootlet bundle near the centre will 
be found to be cut transversely, and then, as stated above, will generally 
appear to have the smallest elements at the centre, surrounded by larger 
elements, possibly metaxylems (see Fig. 4, PI. XV). This mesarch arrange- 
ment of the bundle, due, no doubt, to the inclusion of some of the centripetal 
primary wood of the axis from which the rootlet starts, prevents the bundle 
from having that tongue-shaped appearance in tangential section which 
is so generally characteristic of Stigmaria } where the primary xylem is 
centrifugal. The rootlet bundle is, however, generally connected with the 
secondary wood by special tracheids running down to it, as seen in Fig. 4. 
In these particulars our Stigmaria agrees with that from the Champs des 
Borgis, described by Renault (’ 93 ). In some cases the appearance is more 
normal, and the protoxylem seems to have metaxylem developed only on 
its outer sides (see Fig. 3, PI. XV). The difference in these two arrange- 
ments would, no doubt, depend on the amount of primary wood which takes 
part in the formation of the rootlet. 
On leaving the central woody cylinder, the lateral bundle always shows 
the centric arrangement so characteristic of this plant, and resembling the 
condition found occasionally by Renault in Stigmaria Brardii (see Fig. 5, 
PL XV). Immediately to the outside the bundles show a connexion with 
the wide reticulate cells which occur in large masses, though whether 
continuous in the living condition of the plant and forming an interlacing 
network like those of Stigmaria Brardii cannot be ascertained, owing to the 
defective state of preservation of the cortical layer in which these cells lay. 
By comparison with the similar cells of Stigmaria Brardii it seems likely 
that the reticulate cells formed a network in the cortical region. They 
were evidently connected with the vascular bundles, as can be seen both 
when these latter are cut transversely (Fig. 6 , PL XV) or when a longitudinal 
view of them is obtained (Fig. 7, PL XV). In the latter case the vascular 
bundle is curiously constricted at intervals, not so much by a narrowing of 
the outer reticulate tracheids as by a constriction of the inner elements. 
Similar constrictions are shown in the leaf-trace of the Lepidodendroid stem 
