2 



taken up for the plant which lung passed under the name of Aris- 

 tida disj^ersri T. & R., by Scribner.^ A. fasciculata Torr., is a 

 synonym of this species. A. fasciculata and A. purpurea ha^ve 

 previously been confused, although they are not even closely related. 

 The confusion of these two species doubtless occurred from the fact 

 tlvAt fasciculata was interpreted to apply to habit of growth rather 

 than to the inflorescence. Nearly all of the other species here con- 

 sidered have at one time or another been referred to A. purpurea 

 or A. fasciculata, either as synonjmis or as varieties. A. cequi- 

 ramea Scheele, treated here as a variety of Aristida purpurea^ is 

 perhaps worthy of specific rank. 



In studying this group, one of the chief difficulties has been the 

 large amount of young or imperfectly developed material in the 

 herbarium, which, in many cases, we found impossible to determine 

 satisfactorily ; and, in addition to this, the great variability in habit, 

 doubtless due to environment, added to the difficulty in determining 

 the limits of the several species. It is hoped that the present 

 paper will at least lead to some uniformity in determining the species 

 here considered, as heretofore the names Aristida piupurea, A. 

 fasciculata,, A. fendleriana, and A. longiseta, etc., have been 

 variously applied, first to one form, then to another. 



A. Culms branched, awns not exceeding 2 cm. in length. 



Aristida americana Linn. Amoen. Acad. 5:393. 1759. (Pugill. Jam. PI.) 

 (Aristida fasciculata Torr. Ann. Lye. N. Y. 1 : 154. 1834; Cha'ta)-ia fascicu- 

 lata Schult. Mant. 3:578. 1826; Aristida dispersa Trin. & Rupr. Agrost. 

 3:129. 1842.) 



The original description of this species in Amoenitates Academicae is as follows : 

 " Aristida americana. (j7'a«!cn Festncam referens. Cnlmns ramosus. Differt 

 ab Aristida ascensionis quod radii panicuJa' sint indivisi. Spiculis alternis 

 e flosculis sessilibiis, quornm valvula calycina purpurascens est." 



Munro'^ says regarding the specimen in the Linnean Herbarium: ''Aristida 

 americana L. Am. Acad. 5 : 393. From Jamaica, Browne. This is called 

 .4. disjjersa Trin. \ but Linnteus's name ought to take precedence. Kunth 

 has misplaced the Linnean synonym in Enfriana juncifolia." Linna?us 

 compares Aristida americana with his A. adscensionis, to which the form 

 here considered, is closely related, not only in habit but also in spikelet char- 

 acters. The description of Aristida americana in later editions of Linnaeus's 

 works is very different from the original one given above, and this may 

 explain why Kunth referred this species to Eutriana (Bouteloua). 



The type of Aristida fascicidata Torr., in the Herbarium of Columbia University, 

 bears the label, "Aristida fascicidata Torr., Forests of the Canadian, 

 Long's Expedition, Dr. James," and has been variously interpreted by dif- 

 ferent authorities. This specimen is well matched by No. 66 E. Palmer, 

 Guaymas, Mexico, 1887, distributed as Aristida dispersa Trin. & Rupr. 



' Bui. Torr. Bot. Club 9:87. 1888. 



= On the irlentiflcatlon of the Grasses of Linnaeus's Herharium, Journ. Linn. Soc. Bot. 6:149. 



1862. 



