Holm, Some new anatomical characters for certain Gramineae. 11 1 



arete, simple non cadnque et placke entre deux soies", while this 

 same organ, the arista, is described as being „trifide au sommet, 

 articulee et caduque" in Arthraiherum. His Chaetaria comprises, 

 on the other band, the species in which the flowering gluroe*) 

 is „plus ou moins prolongle en pointe, terminee par trois soies^ 

 le plus souvent egales", while the llowering glume in Curtopogon 

 is described as „fendue, bilacinee, et une seule soie tordue entre 

 les dents . . .". Hence Beauvais drew a distiuction between 

 pSeta" and „arista", the latter being characteristic oi Arthratherum 

 and Aristida ; this organ is articulated and deeiduous in the former^ 

 but not so in the latter. 



Bentham and Hook er receive the genus in its old extent^ 

 but recoguize as subgenera or seetions Beauvais' genera : 

 Chaetaria and Arthratherum, besides Stipagrostis, which was origi- 

 nally established by Nees as a proper genus. CJiaetaria is by 

 Bentham and Hooker an aggregate of both Beauvais' 

 Chaetaria and Curtopogon besides that it includes Streptachne H. 

 B. K. The s>eciion Arthratherum is with Bentham and Hooker 

 only a part of Beauvais' genus, embracmg the species with 

 deeiduous, but naked awns, while their section Stipagrostis com- 

 prises all those species in which the awns are plumose, namely 

 Arthratherum (pungens) and Aristida (ianata) both of Beauvais. 



A similar disposal is suggested by Ha ekel in his treatment 

 of the family for' Engler and Prantl's „Natürliche Pflanzen- 

 familien", and so is „CÄaetorm" as a section designated to the. 

 species having naked, not deeiduous awns: ^Arthratherum'^ to 

 those with naked and deeiduous awns, while .^Stipagrostis'^ includes 

 the remainiug in which these organs are deeiduous and plumose. 

 The genus-name ^^Aristida'^ as taken up by Beauvais only for 

 A. Ianata Forsk., which according to Vahl**) is identical with 

 A. plumosa L., has not been adopted in this wise by subsequent 

 authors, but as the name of the whole genus, including the seetions, 

 enumerated above. 



Neither has Nees von Esenbeck's ^^ Stipagrostis'^ been 

 restricted to comprise those certain species, which he described 

 in Agrostographia capensis***) but has been extended so as to include 

 also a part of Beauvais' Arthratherum with plumose awns, 

 although Nees von Esenbeck himself considered Aristida 

 plumosa L. and ^. ciliata Desf. as representatives of Arthratherzim^ 

 but not of Stipagrostis. That Nees did not recognize A. plumosa 

 L. as an „Aristida'-^ in the sense of Beau/ais was evidently in 

 view of the fact that neither the diagnosis or the Illustration in 

 Beauvais' work are correct in this particular case : the arti- 

 culation of the awn seems to have been overlooked, and so 



*) The term „empty glumes" is used here for „glumae vacuae" and 

 „flowering glume" for „gluma florens" (Bentham and Hooker: Genera 

 plant.)- 



**) Vahl, Martin: Symbolae botanicae. Pars prima. Kjöbenhavn 1790. 

 p. 11. 



***) Halle 1853. p. 171. 



