20 



for distinguislimg tliem by anatomical characters, yet it is interesting 

 to note that in case of such a contingency separation of the species by 

 structural peculiarities would be an easy matter. It would be of impor- 

 tance that the leaf to be examined should be fully developed, and if only 

 a portion of a leaf were at hand to know whether it were basal, median, 

 or apical. The latter question is not of vital importance, however, 

 though where there is a choice in the matter a median section is best. 

 The staminate plants of J. inlosa and J. straminea coukl not be sepa- 

 rated by anatomical characters, but assuming as the standard a median 

 section of a full-grown leaf of a 9 plant of each species, the following 

 table would serve to separate these four grasses : 



Leaves unfurrowed on the upper face 9 Joitvea straminea. 



Leaves furrowed ou the upper face. 



1. Colorless parencbyuia below the bypodermal fiber at the summit of the ridges, 



Eragrostis odttinflora. 



2. No colorless parenchyma at the summit of the ridges. 



a. Leaf thick; inferior epidermis smooth or nearly so; fiber situated under the 



colorhiss parenchyma between the bundles Jouvea pUosa. 



b. Leaf thin; inferior epidermis rough with numerous expansions; no strand 



of fiber under the colorless parenchyma between the bundles, 



iJistichlis sjncata. 



