
36 BOTANICAL GAZETTE [JANUARY 
considered as inseparable, while the inner sheath showed some 
resistance wherever it occurred. 
CONCLUSIONS. 
In comparing the general structure of £. decangulare with 
that of &. helichrysoides, as described by Poulsen, there seem to 
be certain peculiarities in common which may be characteristic 
of the order, inasmuch as they have also been observed in some 
of the other genera. The presence of collenchyma is one of 
these, although this tissue should not be considered so rare in 
monocotyledonous orders as Poulsen claims it to be. It is com- 
mon, for instance, among the Graminez, where it is well devel- 
oped in the nodes; it is also foundin the small “ pulvini’” which 
occur at the base of the lateral branches in paniculate inflores- 
cences of a number of genera. Moreover, in the Cyperacee 
collenchyma occurs in the basal portion of both rachis and pro- 
phyllum in certain species of Cyperus of the subgenera Eucype- 
rus, Diclidium, and Mariscus; besides in such species of Scirpus 
as possess a large open inflorescence (S. atrovirens, S. polyphyl- 
‘ dus, and others). A still more advanced development of this 
tissue may be observed in Commelinacee, Smilacez, and Dios- 
coreacez. It does not seem, therefore, as if this tissue exists 
only in ‘‘a very few monocotyledons.”’ But it seems charac- 
teristic of the Eriocaulacee examined that the collenchyma, or 
to be more exact ‘‘collenchymatic tissue,” occurs as prominent 
ridges in the stem, that it replaces the stereome in the leaves, 
and that it surrounds the mestome bundles as a closed sheath. 
The inner mechanical sheath, noted in the leaves, appears to have 
several points in common with the mestome sheath of the Gram- 
inez, yet it does not seem to be identical with this, and perhaps 
may be characteristic of Eriocaulon. The hairy rhizome consti- 
tutes an excellent character of the order, even if a few instances 
are known in other orders of hairs on underground stems, Viz 
Corallorhiza, Scheuchzeria, and Mercurialis. Otherwise the 
structure of Eriocaulon does not differ so very much from that 
of certain other orders, as Cyperacee and Juncacez, in which a 

naan sanescanenrmacaremnienmnanmasiaanasnitite is neal re | 
en 

