1907-8.] Origin of adaxially-curved Leaf-trace in Filicales. 
433 
XXIX. — On the Origin of the adaxially-curved Leaf-trace in the 
Filicales. By D. T. Gwynne- Vaughan, M.A., and R. Kidston, 
LL.D., F.R.S.L. & E. 
(Read May 4, 1908. MS. received same date.) 
While studying the fossil Osmundacece several interesting points illus- 
trating the origin of the leaf -trace were brought under our notice, which are 
exhibited with such exceptional clearness in Thamnopteris Schlechtendalii, 
Eichwald, sp.,* that they deserve especial recognition. The full discussion 
of the subject, however, is reserved until the publication of the next part of 
our memoir on the fossil Osmundacece. 
The departure of the leaf-trace in Thamnopteris is typically protostelic. 
That is to say, the xylem of the leaf -trace first of all appears as a protuber- 
ance on the surface of the xylem of the stem (fig. 1). Later, this separates 
off without leaving any depression or gap in the stem xylem. Immediately 
after its departure the leaf-trace xylem is elliptic or oblong in transverse 
section, with a single mesarch protoxylem group, almost exactly in the 
middle of the strand (fig. 2). Further out, certain of the centripetal 
tracheae situated in front of the protoxylem on the adaxial side of the 
trace cease to be formed, their place being taken by cells of thin-walled 
parenchyma (fig. 3). An isolated island of parenchyma is thus produced in the 
xylem strand just in front of the protoxylem group. This island gradually 
increases in size in the outer traces, more and more of the tracheae on the 
adaxial side being suppressed, until at last it opens out to form an adaxial 
bay of parenchyma (fig. 4). For some time a few centripetal tracheae 
are still to be found in immediate contact with the actual protoxylem 
elements, but these also eventually die out, and the protoxylems become 
truly endarch. 
As the leaf-trace passes further out through the inner cortex of the 
stem the xylem becomes tangentially elongated, and the bay of parenchyma 
progressively wider and more open until the xylem strand assumes the form 
of a stout crescent with thick, incurved ends. At the same time the proto- 
xylem elements spread out over the concave margin of the bay (fig. 5), and 
* Eichwald, Lethcea Rossica , vol. i., p. 93, pi. xx., figs. 2 and 5, 1860 ; Brongniart, 
Tableaux des genres des ve'getaux fossiles, pp. 35-36, 1849. 
VOL. XXVIII. 
28 
