570 Weiss. — -The Vascular Branches of 
tissue to the various protoxylem groups. The same may be 
said of most recent aquatics. The only existing type of 
aquatic with a monarch root, lsoetes ) has as a submerged 
plant a much smaller requirement for root-absorption, and 
is apparently able to dispense with this additional method of 
conduction shown to be characteristic of the various types of 
Stigmarian rootlets. 
General Conclusions. 
In the various types of Stigmarian rootlets fine strands of 
spiral tracheids, like those described by Renault, may be seen 
leaving the protoxylem elements. 
These strands do not pass out to lateral rootlets as suggested 
by Renault, but terminate in the outer cortex, sometimes in 
connexion with distinct groups of large parenchymatous cells. 
The vascular strands are not directly connected with the 
parenchyma of the outer cortex, but pass out into short and 
wide spirally-marked cells resembling the transfusion cells of 
leaves. 
The vascular strand and the transfusion cells in which it ter- 
minates form a special means of conducting water from the 
peripheral to the central tissues of the rootlet, a means which 
is rendered necessary by the development of the middle cortex 
into an air-conducting tissue or space. 
The existence in the cortex of Stigmarian rootlets of spiral 
tracheids comparable with the transfusion-cells of Lepidoden- 
droid leaves may possibly be taken by some palaeobotanists 
as a further argument in favour of Schimper’s view of the 
homology of these two sets of organs 1 . 
In making the above comparison, however, I desired merely 
to refer to them as organs of similar structure and analogous 
to some extent even in function, in so far as they establish 
a passage between the parenchymatous and vascular tissues. 
I have searched in vain in Lepidodendroid leaves for the fine 
vascular strand which traverses the middle cortex of the 
1 Schimper, W. Ph., Pal^ontologie v^getale, 1872, vol. ii, p. hi. 
