1906] Lamson-Scribner, — Trisetum and Graphephorum 85 



rachilla, which he makes the essential character of his new genus, 

 placing it with Triodia. 



Beauvois in 1S12 recognizes Graphephorum, and adds a little to the 

 generic characters without increasing the number of species. He says 

 (Agros. 70): "Flosc. suprem. pedicellato, abortivo, villosissimo. — Pa- 

 leae bifido-dentatae: infer, inter dentes mucronata." He evidently 

 looked upon the hairy prolongation of the rachilla as an aborted floret, 

 and his specimens showed the minute awn below the teeth of the 

 flowering glume, which character w^as not manifest in the material 

 examined by Michaux and Desvaux, or was overlooked by them, but 

 which is nevertheless common, the awn often being quite conspicuous. 

 In 1856, Dr. Asa Gray published (Man. ed. 2: 556) as new a form 

 of Aira meUcoides Mich, under the genus Dupontia, naming it D. 

 Cooleyi and comparing it with Aira racspitom and Aira botknica. 

 Later, having discovered the relation of his grass with Michaux's 

 Aira inelicoides, Graphephorum melicoideum of Desvaux, he revised 

 the genus Graphephorum (Annal. Bot. Soc. (\anada, 1': 55-57, 1861) 

 modifying its characters so as to include the species of Scolochloa, 

 Dupontia, and Colpodium, reducing his Dupontia Cooleyi to a variety 

 of Graphephorum meliroide.s', characterizing it as "a luxuriant form 

 from 2 to 3 feet high with ampler panicles." 



There is in the Gray Herl)arium a specimen collected by Dr. Cooley 

 in Macomb Co., Michigan, which is doubtless the type of this variety, 

 and in its robust habit and pilose leaves. Gray very naturally failed at 

 first to connect it \\\i\\ the more slender and glabrous plant described 

 by Michaux. This specimen has scabrous leaves which are pilose 

 upon the up])er surface, spikelets 6-7 mm. long, unequal empty glumes, 

 the broad ;i-nerved second glume nearly equalling the spikelet and the 

 oblong obtuse flowering glumes, which are entirely awnless. The 

 rather stiff hairs on the rachilla are about 1.5 mm. long. This variety 

 is exactly represented in the National Herbarium by 26,222 J. Macoun, 

 from Johnston's Harbor, Lake Huron, collected in 1901. 



Both Hackel (Engl. & Pr. Xaturl. Pflanzenf. 2^: 74, 1887) and 

 Baillon (Hist. d. Plant. 12: 212, 1894) hold Graphephorum as a genus 

 distinct from Dupontia, Scolochloa, and Colpodium; but regard it 

 as being closely allied to them, placing it with them in the Festureae. 

 This disposition appears to be wholly unwarranted, for there is nothing 

 in common between Graphephorum and the genera above named ex- 

 cepting the hairiness about the base of the flowering glume, while there 



