686 
Kershazv.—A Fossil Solenostelic Fern . 
description, and it is reasonable to assume that the whole of the rhizome was 
of this solenostelic type. 
The xylem is composed of a ring of tracheides, one or two layers deep, 
in the ventral portion of the stele (Figs, i, 2, and 3,#.), but becoming five or 
six layers deep in the dorsal portion in connexion with the outgoing leaf- 
trace (Figs. 1, 3, and 4, a). A few parenchymatous cells occur here and 
there among the woody elements. Protoxylem elements of the spiral and 
annular type are not found in the stem, but fairly frequent exarch groups ot 
scalariform elements are found around the ring of metaxylem ( p.x ., Fig. 2), 
quite similar to the ‘ protoxylem elements * described for recent solenostelic 
ferns. 1 Unfortunately there are no longitudinal sections of the rhizome, but 
in some of the oblique parts of the transverse sections it may be seen that 
both the protoxylem and metaxylem are composed of scalariform elements, 
which possess a single series of pits on each wall. In some of the better- 
preserved and uncrushed parts of the rhizome distinct pits can be seen 
(Fig. 5) between the walls of adjacent xylem elements, such as Gwynne- 
Vaughan 2 has described in the fossil Osmundaceae, and has since shown to 
exist in the xylem elements of many of the living ferns. 
The phloem forms a continuous ring on either side of the xylem (Fig. 2, 
i.ph . and o.ph .), but unfortunately, as is the case with the other soft tissues 
in the stem, it is usually so crushed that the structure is not very clearly 
seen. In the better-preserved parts one can recognize two or three layers 
of sieve-tubes, but no distinction into protophloem and metaphloem. The 
phloem layer is separated from the xylem on each side by an irregular 
layer of parenchymatous cells (Fig. 2, par.). Surrounding both the inner 
and outer bands of phloem is a single layer of pericycle cells (Fig. 2, i.per. 
and o.per .), often rather irregularly arranged. The solenostele is limited on 
each side by a well-marked endodermis, which is clearly shown in all 
sections, since the cells, unlike the surrounding ones, are usually filled with 
masses of black, carbonized material. They are slightly elongate tangentially 
—that diameter being 0-02 mm. (Fig. 2, o.end, and i.end .). There are no 
undulations on the radial walls of the cells, but the corners are slightly 
thickened. This endodermal layer is remarkably well defined and recog¬ 
nizable for a fossil. 
The Leaf-trace and Branch. 
As was mentioned previously and shown in the Text-fig., the continuity 
of the solenostele is interrupted in some sections by the outgoing leaf- 
trace. The lateral shoot of the rhizome, as in living solenostelic ferns, 
1 Gwynne-Vaughan (’ 03 ), Observations on the Anatomy of Solenostetic Ferns (Part II). Annals 
of Botany, vol. xvii, 1903. 
2 Gwynne-Vaughan (’ 08 ), On the real nature of the Tracheae in the Ferns. Annals of Botany, 
vol. xxiii, July, 1908. 
