1914] PFEIFFER—THISMIA AMERICANA 127 
appears brown. The septate hyphae are usually oriented with the 
long axis of the root, so that the cross-section of a root shows 
numerous cut ends (fig. 14), and the longitudinal section a more 
or less parallel arrangement of the interweaving hyphae (fig. 7). 
Below this single layer is a region of a varying number of cortical 
cells containing much finer, thin-walled hyphae. In these deeper 
parts, masses of protoplasm are frequently evident which strongly 
suggest the sex organs of some of the Peronosporales. These 
undoubtedly correspond to the ‘vesicles’ reported by JANSE 
(24) in Thismia clandestina (T. javanica J. J. Sm.), which he was 
inclined to believe asexually reproductive bodies. BRUCHMANN 
considered similar bodies in Lycopodium annotinum to be oospores 
of Pythium. In Thismia there are also bacteria, probably corre- 
sponding to JANSE’s “sporangioles” and ‘“‘spherules.’”’ All these. 
fungal parts are intracellular. A few cells outside of the endodermis 
are free from fungi. These contain raphides which are common 
throughout the plant body. 
The endodermis consists of a single layer of heavy-walled cells, 
larger than their neighbors. It encircles a region, probably con- 
ducting, of which only a few central cells (3-5) are spirally thickened 
(figs. 5, 6, 19). They are not lignified, however, and the spiral 
markings are very fine. Near the point of origin of a floral axis, 
the number of thickened cells is increased and lignification occurs. 
These vessels may be seen to connect directly with the vascular 
elements of the floral axis. The cells adjacent to these spirally 
thickened vessels do not show the dotted condition that Jonow 
Teported in the Burmanniaceae considered by him. On the con- 
trary, though they are somewhat elongated, they are nucleate and 
retain their cytoplasm. They are undoubtedly parenchyma cells, 
and so the situation is similar to that in Thismia javanica J. J. Sm. 
and other forms investigated by ERNST and BERNARD. 
In the outer part of the conducting region are seen a varying 
number of points of small cells devoid of contents (fig. 6). The 
arrangement is similar to that reported in other forms by JoHow 
and by Ernst and Bernarp. It suggests a radial arrangement, 
with these groups of cells probably reduced phloem strands without 
Sleve plates. 
