70 Sykes . — The Anatomy and Morphology of Tmesipteris. 
has the following structure (Fig. 2, PI. VII). The epidermis is thin-walled 
and has no stomata ; the cortex is collenchymatous, and its innermost layer 
is impregnated with a brown substance ; no clear endodermis can be 
distinguished, but the continuous layer of phloem is separated from the 
cortex by some thin-walled, non-pitted, cells, which include perhaps both 
endodermis and pericycle. The phloem is itself made up of elongated ele- 
ments, the walls of which are partly lignified and are scattered with numerous 
lateral sieve plates. On staining with Iodine Green and Eosin, the general 
wall of a sieve-tube stains green, and the bright pink sieve-areas appear 
very prominent. This curious bonification of the sieve-tubes in Tmesipteris 
may be compared with that described in Helianthus 1 , and with the sclerosed 
sieve tubes of Loxsoma 2 , Schizaeaceae 3 , Gleicheniaceae and Trichomanes 
Prieurii 4 . It was early observed by Russow 5 in Tmesipteris , but his state- 
ment was negatived by Jennings and Hall 6 . Internal to the phloem, a ring 
of five or six mesarch xylem strands surrounds the so-called pith. Each strand 
is composed of two or three half lignified, finely scalariform, protoxylem ele- 
ments, enclosed by broadly scalariform metaxylem tracheides. This arrange- 
ment of separate mesarch xylem strands, surrounded by a continuous layer 
of phloem and enclosing a parenchymatous pith, recalls the structure of 
Osmunda and Todea. The elongated pith-cells are pitted on their 
tangential walls, and are often collenchymatous ; they contain proteid 
masses such as were above described in the phloem. Jennings and Flail call 
these elements sieve parenchyma because of their profusely pitted walls. 6 
The course of the xylem strands is generally straight, with very little 
anastomosis in the lower part of the stem, but after we reach the leaf-trace 
region we find a good deal of irregular fusion taking place between them. 
(d) Origin of Leaf-Trace. The ordinary leaf -trace arises in the 
following manner. One of the protoxylem strands of the stem divides 
about half a centimetre below the departure of the leaf-trace. The outer 
of the two products of division becomes separated from the inner, and 
gradually forms an independent group, enclosed by metaxylem. This 
group projects more and more into the phloem and finally separates 
from the xylem, becoming in its passage outwards surrounded by phloem, 
pericycle, and a layer of brown cells (Text-fig. VI, lt 3 ). Thus it will 
be seen that the leaf-trace leaves as a rule no leaf-gap. The first leaf- 
trace often differs slightly in its formation from the later ones ; the division 
of the protoxylem of the group which will give rise to it takes place rather 
sooner than is usual, and the xylem of the trace is then nipped off and 
continues for some time as a separate group in the stele. 
1 Boodle, 1902. 2 G wynne Vaughan, 1901, pp. 84-5, PI. Ill, Fig. 10. 
3 Boodle, 1901, pp. 400, 417, PI. XX, Fig. 19; xxi, Fig. 46. 
4 Boodle, 1901, p. 714, and Poirault, 1 . c., p. 190, 195. 
5 Russow, l.c., p. 132. 0 Jennings and Hall, 1891. 
