Holzinger : ON THE GENUS COSCINODOX. 757 



of the more naked lower part of the shorter stem, the leaves 

 being more crowded toward the upper end. I began to think 

 this might be a third species of Coscinodon, after all. But when 

 I had looked up Cardot's letter, above quoted, in which he 

 insists that this plant is Coscinodon wrightii, I was again at sea. 

 And now I determined upon a critical examination of leaf- 

 cross-sections, the result of which leaves no doubt regarding 

 the relationship of our Minnesota Coscinodons, furnishing also 

 absolutely reliable diagnostic characters of the two species. 



I first proceeded to make leaf sections of the two Colorado 

 plants, with the following results: 



1. Coscinodon lurightii yielded leaf sections uniformly and 



evenly concave, in which the costa projected nearly 

 evenly above and below the leaf surface; toward the 

 apex the costa is restricted to the outer, under side of 

 the leaf; in both sections a single row of large cells 

 lies across the upper surface, continuous with the leaf 

 cells. See PL XLI, Jig. S. 



2. Coscinodon raid has a deeply furrowed costa, as is shown 



by Jig. 12. This form of the costa doubtless gives 

 the rigid divergent appearance to the stem leaves of 

 G. raui under the hand lens. That the furrow extends 

 even some distance into the colorless awn of this 

 species is made apparent under the low power of the 

 microscope by the occasional accumulation of soil in 

 this groove, a condition frequently observed in this 

 plant. Such an appearance never occurs in C. wrightii, 

 which, judging from the leaf sections, has probably 

 no such groove in the hair points of its leaves. 



With this diagnosis worked out, I made a similar examina- 

 tion of cross sections of the Minnesota plant. And, judging 

 from this character, there can be no doubt but that it should 

 be referred to C. wrightii rather than to C. raid, its leaf sec- 

 tions being exactly like those of C ivrightii from Colorado. 

 But there was still a recognizable difference in the gross ap- 

 pearance of my Minnesota plant, a difference uniform in all my 

 material, and which I established by a series of examinations. 

 This, in my judgment, justifies the Minnesota plant to be des- 

 ignated as a variety of C. wrightii, for which I propose the 

 name — 



Coscinodon wrightii var. brevis n. var. 



