258 
Birbal Sahni. 
is practically uniform in thickness all round. In the sieve-tubes the 
refringent granules are more numerous and the protoxylem groups 
often fewer by one or two than in the primary stolon. 
The mode of origin of the secondary stolon 1 is illustrated by 
the diagram in Text-fig. 3. The branch strand is thin to start with, 
and runs for a couple of millimetres nearly parallel to the vascular 
cylinder of the primary stolon and still enclosed in the cortex of the 
latter, at the same time gradually widening out. Then it sharply 
turns away almost at right-angles to its previous course, forming a 
a striking knee-like bend. Attention may here be called to the 
resemblance of this mode of branching to the origin of the so-called 
“ axillary branches ” in the Hymenophyllaceae and the Botryopter- 
ideae. A comparison of Text-fig. 3 with Boodle’s 2 figure of a 
longitudinal section through the region of branching in Trichomanes 
radicans is interesting, even if it only brings out a superficial 
resemblance. 
At the actual point of branching, forked or V-shaped tracheids 
(all scalariform) were seen, similar to those observed by Boodle 3 in 
Lygodium. The tracheids in the ascending part of the branch-strand 
are very short and more or less distorted. 
The behaviour of the protoxylem strands is interesting. Text- 
fig. 4 shows diagrams drawn from steles actually dissected out, and 
based on several specimens which could be reduced to the three 
types, b, c, d. In this figure, a represents a side-view of b. The 
preparation for the branching of the stolon is often apparent as 
much as half-a-centimetre below the point where the stele forks. 
One, two or three of the protoxylem-strands of the primary stolon, 
which are destined to supply the secondary stolon, pass upwards 
and divide as shown (thicker lines) the branches gradually spreading 
round the metaxylem cylinder of the secondary stolon to the extent 
that the latter becomes free from the main. The median protoxylem 
is entirely given over to the branch-supply, but each supplementary 
one, when present, partly supplies the branch, and is partly 
continued into the main. The number of protoxylem-groups in 
the main stolon, above and below the point of branching remains 
1 Though Sperlich (l.c., p. 462) goes into the branching of the stolons 
in N. cordifolia, the relations of the vascular structures have not been 
elucidated. 
J Boodle, Ann. of Bot., 1900, fig. 28, pi. xxvii. 
* Boodle, Ann. of Bot., 1901, pp. 367-8. 
In one preparation of N. rufesceus var. Whitmanni two or three V-shaped 
tracheids were seen fitted together in a row, while in another one two curved 
sieve-tubes were observed similarly arranged. 
