66 9 
certain Orders of the Ranales. 
From the above it will be seen that each leaf, besides the large midrib- 
strand, receives four bundles (two on each side) from the cauline inverted 
strands of the stem. 
Essentially the same facts were observed in the case of the seedling 
plant ; it would appear, however, that the cauline bundles are rather late 
in making their appearance, for in the region where the cotyledonary 
bundles have left the cylinder, but where the cotyledons themselves are not 
yet detached as distinct organs, there is no sign of the four cauline bundles. 
But these latter soon after arise from the cotyledonary traces ; or, in other 
words, they pass downwards from the first pair of foliage-leaves above and 
unite with these traces. 
The facts above detailed are usually interpreted as follows : that 
each cortical bundle is built up by the lateral bundles of one-half of each 
leaf (the main midrib-bundle passing directly into the cylinder), which on 
entering the stem at once unite with a similar leaf-trace descending from 
the node above. These ‘ leaf-traces then, never unite with the stem- 
cylinder at all, but constitute an independent cortical system. As was the 
case also with Lignier and Van Tieghem, I was unable to discover any 
nodal connexion between the cortical bundle and the stem-cylinder, such 
as Woronin describes ; this connexion is effected solely by the midrib- 
bundles. 
Baillon describes a Chimonanthus-sXem whose leaves (except those, 
as in the case of some shoots, at the base, which were opposite) were 
arranged on the shoot according to a f -spiral, while, in accordance here- 
with, there were five cortical bundles ; this may be regarded as of the 
nature of a reversion to the primitive condition, for it is interesting to note 
that the peduncle of both genera, which bears spirally-a.rrd.ngQd foliar 
organs, has from six to eight cortical bundles which are in five groups 
(i. e. at two points a pair of bundles occurs). The probable physiological 
explanation of the presence of these persistent and independent ‘leaf-traces’ 
which I would advance is that they serve to give rigidity at the four necessary 
points to the angles of the quadrangular stem, and have been preserved for 
this purpose ; the same phenomenon of cortical strands occurring in such 
a position is seen in the square stems of Labiatae, Melastomaceae, and other 
plants. 
Bibliography. 
Baillon : Sur un Chimonanthns a feuilles alternes. Adansonia, vol. ix, 1868-70. 
Cruger : Anomale Holzbildungen. Bot. Zeit., p. 481, 1851. 
De Bary : loc. cit. 
Gaudichaud : Arch, de Bot., vol. ii, 1833. 
Hkrail : Recherches sur l’anatomie comparee des tiges chez les Dicotyledones.’ Ann. Sci. Nat., 
ser. 7., vol. ii, 1888. 
