Igo BOTANICAL GAZETTE [SEPTEMBER 
detect any character that seems peculiar to these sections; on the 
contrary, the rhizome, the leaves, and the inflorescence are very 
uniformly developed, and the only distinction depends upon the 
floral structures which were mentioned in the introduction. If these 
floral structures were accompanied by differences in the development 
of the shoot, and also by peculiarities in the internal structure, there 
might have been some reason for dividing the genus. There are 
several very peculiar looking species in Mexico, and these ought to 
be examined anatomically from fresh material. Until then it seems 
most advisable to leave the genus intact. 
Finally, I wish to add a few words in reference to the structure 
of the chlorenchyma in other genera of the Irideae. We have seen 
the chlorenchyma in the sections BERMUDIANA (fig. 19) and ERI- 
PHILEMA is more or less differentiated as palisades, while in ECH- 
THRONEMA this tissue is composed of oblong cells parallel with 
epidermis; moreover, that the cells in ECHTHRONEMA are lobed, 
resembling those of a pneumatic tissue. These structures occur also 
in other genera, and for the sake of making a comparison I examined 
some of them. In Iris, for instance, we find in J. cristata Ait. (fig. 22) 
a structure of the chlorenchyma which corresponds well with that 
of Sisyrinchium californicum, but the cells of which are branched 
rather than lobed, thus representing a typical pneumatic tissue. In 
Iris verna L. and I. julva Ker., on the other hand, there are typical 
palisades, which, however, show the same direction, being parallel with 
the epidermis and vertical to the longitudinal axis of the blade; ™ 
I. xiphioides Ehrh. the palisades are vertical to the epidermis. — 
structure that corresponds with that of Iris verna is furthermore 
characteristic of Tapeinia magellanica Juss., Freesia rejracta Klatt., 
and Tritonia sp. In Belamcanda chinensis Adans. the chloren- 
chyma has a very irregular structure, the cells varying from oblong 
and entire to more or less deeply lobed. We have thus in Ins * 
varied a development of the chlorenchyma as was observed in Sisyt™ 
chium. 
I have also pointed out the somewhat peculiar arrangement - 
the mestome strands in Sisyrinchium (the single plane in BERMU- 
DIANA, and the narrow elliptical band in Errxmema and : 
THRONEMA). That such variation in the position of the mestome 
