TRANSACTIONS OF SECTION K. 665 
of the ventral canal cell, which in many cases could not be demonstrated 
at all. 
The embryo becomes elongated in the direction of the archegonium axis 
before division; and at this time resembles that of Botrychium obliquum. 
The first division wall is transverse as in other Marattiaces, but the 
hypobasal cell either does not divide at all or divides only once and forms 
a short suspensor, all of the organs of the embryo arising from the inner 
or epibasal cell, The latter undergoes a somewhat irregular quadrant 
division, the two lower quadrants forming the foot, the upper ones giving 
rise to the stem apex, the leaf, and later the root. . 
A single large apical cell can usually be demonstrated in the stem 
apex at a very early period. No single initial is present in the young 
leaf, which does not appear to be always formed at the same point. 
No trace of the root can be made out until the embryo has reached a 
considerable size. The root is strictly endogenous in origin, and its single 
initial cell arises nearly in the centre of the embryo, probably from the 
stem quadrant. With the elongation of the root downward, it carries with 
it the foot which covers the growing point of the root like a root-cap. 
No cauline bundle is developed in the young sporophyte, the vascular 
cylinder being made up of the coalescent leaf-traces. 
3. On the Ancestry of the Osmundacee. By R. Kipston, LL.D., 
F.R.S., and Professor D. T. Gwynnn-Vauauan, M.A. 
Lalesskya gracilis and diplorylon and Thamnopteris schlechtendahlii, 
from the Permian of Russia, are regarded by the authors as primitive 
Osmundacez. Their conclusions are based upon the structure of the stems 
and leaf-bases which alone are at present known. 
The vascular system of the stem is a protostele with a solid mass of xylem. 
The xylem is not homogeneous, however, but consists of a stout peripheral 
zone of normal scalariform trachee, surrounding a central mass of short 
swollen tracheal elements, with porose pitting and probably with more or less 
thin walls. The petiolar meristeles are C-shaped with the concavity facing 
towards the axis, and they show a very close resemblance to those of the living 
Osmundacee, even in detail. 
It is held that the pith of the modern Osmundacee is a true pith, and that 
it arose by the transformation of the above-mentioned central xylem into thin- 
walled parenchyma. Later the peripheral xylem ring lost its continuity and 
became broken up into the separate strands characteristic of the steles of the 
existing forms. 
The Osmundacez and the Zygopteridee are held to be derived from a 
common ancestor. The vascular system of their stems exhibits a parallel series 
of developments in the two orders. The typical Zygopterid stele of the 
Upper Carboniferous has reached a point in advance of that possessed by 
Thamnopteris in so far that most of the central trachese have been replaced by 
parenchyma, although some still retain their tracheal characters, but those of 
the Lower Carboniferous may be expected to exhibit a stele essentially identical 
with that of Lalesskya and Thamnopteris. 
4. Preliminary Note on the Structure of a new Zygopteris, from 
Pettycur, Fife. By W.T. Gorvon, M.A., B.Sc. 
Dr. Kidston and Prof. Gwynne-Vaughan have demonstrated a develop- 
ment, among the fossil Osmundace, from a protostelic form to forms 
which have a well-marked, parenchymatous pith. The Zygopterid stems 
so far described have either a mixed or a true pith, but no protostelic form 
has yet been recorded. This new Zygopteris, to which I have given, pro- 
