250 MR. A. C. SEWARD AND MISS 8. O. FORD ON 
confined to one layer, and the cells are larger and wider than those of the better marked 
outer endodermis. 
The leaf-traces (figs. 44, 46, 47, 49, PI. 30) are given off as in the other species, and, 
as in T. superba, the roots do not arise at the same level, nor do they appear always at 
the same stage on the outgoing leaf-trace. In the photograph reproduced in Pl. 30. 
fig. 46, the endodermis of the stele is seen to be bent round the leaf-trace //, and is about to 
separate itself from that of the foliar bundle. Another feature worth noting is the greater 
difference in the diameter of the outer tracheids of the stem as compared with the more 
internal xylem elements (figs. 44, 46, 49, Pl. 30): this is seen also in Z. superba (Pl. 30, 
fig. 41), but is less marked in T. barbara. 
GEOLOGICAL HISTORY. 
As a result of the more minute investigation of Palæozoic plants carried out 
during the last few decades, facts have been gradually accumulating which enable us 
io recognize in the morphological features of many extinct species signs of union in 
the remote past of types of vegetation that are now clearly distinguished one from 
another. The numerous types of fern sori and sporangia, and the various anatomical 
features exhibited by the structure of the vegetative organs of Ferns and Fern-allies, 
afford striking illustrations of the importance of Paleozoic Botany as a guide to 
phylogeny. Existing Ferns are in the main divided into fairly well-defined families 
characterized by clearly marked features; but among Palæozoic representatives of the 
Filicales and Cycadofilices we find, instead of the separate lines of affinity exemplified by 
living forms, an intricate plexus of branches which are not amenable to the same kind of 
classificatory treatment as the modern Ferns. While the majority of the sporangia 
found on fertile pinnze of Palseozoic fronds agree more closely with those of the recent 
Marattiaceze than with the Leptosporangiate ferns, there are several instances of the 
admixture with the Marattiaceous features of others suggestive of a closer union between 
fern families than exists at the present day. The Botryopteride:e represent a family of 
Palwozoie Ferns which cannot be accommodated in any existing division; they afford 
links on the one hand with members of the Osmundace, and on the other hand with 
certain species of Hymenophyllaceze. The sporangia of Botryopteris have been described 
by Renault *, and more recently by Scott; as the latter author points out, the charac- 
teristic and prominent annulus may be compared with the so-called annulus of Osmunda 
or Todea in an exaggerated form. A recently discovered type, described by Zeiller { 
from Asia Minor as Kidstonia, possesses sporangia with a prominent group of cells 
forming a cap-like annulus (Pl. 27. figs. 6 & 9), intermediate in structure between the 
apical aunulus of the Schizæaceæ and the intra-apical group in the Osmundaceæ. 
From the Culm of Esnost, Renault has recorded pyriform and short-stalked sporangia 
under the name Zodeopsis primeva §, characterized by the possession of a group of 
* Renault (1896), pp. 33 et seg. 
t Bec Scott (1890), pp. 297 et seq. 
= Zeller (1899), p. 22. dei : 
3 Renault (1896), p. 21, fig. 18, 
