128 Chase — Notes on Genera of Panicese. IV. 



Steudel (Syn. PI. Grlum. 1 : 56. 1854) follows Trinius' own later limita- 

 tion' of the subgenus Brachiaria.* 



Bentham (Benth. & Hook. Gen. PI. 3:1102. 1883) uses the name 

 Brachiaria for a section of Panicum and in about the sense equivalent to 

 the group Geminata (Hitchc. & Chase, Contr. Nat. Herb. 15 : 30. 1910). 

 Hackel (Engler & Prantl. Pflanzenf. 2 : 2 35. 1887) also uses it as a sec- 

 tion of Panicum and apparently in the same way, his diagnosis being: 

 Inflorescence as in Paspalum, spikelets awnless. 



Nash (Britton, Man. 77. 1901 ) recognizes Brachiaria as a genus (giving 

 Ledebour as the author), differentiating it by the racemose inflorescence 

 and awnless spikelets of 3 glumes. Two species, Panicum digitarioides 

 (P. hemitomon Schult. ) and P. obtusum H. B. K., in both, of which the 

 spikelets are in the position normal for Panic urn and Paspalum, not the 

 reversed position of Brachiaria eruciformis , are transferred to it. Later 

 (in Small, Fl. Southeast. U. S. 50 and 80. 1903) Nash adds to his diag- 

 nosis of the genus Brachiaria "flowering scale with its opening toward 

 the rachis," thus limiting the genus to Grisebach's type species and its 

 allies. Panicum digitarioides and P. obtusum he here restores to Panicum. 

 Hitchcock (Contr. Nat. Herb. 12 : 141. 1908) accepts Brachiaria in this 

 emended sense, separating it from Panicum chiefly on account of its 

 " having spikelets so placed that the fertile floret stands with its palea 

 toward the axis." 



Milium Bubani, Fl. Pyren. 4 : 259. 1901, not L. 1753. This includes a 

 single species, M. altemans Bubani, based on Panicum eruciforme Smith. 

 No generic description is given but something of the author's concept of 

 the genus is shown by his statement that it is not possible to separate 

 Panicum beckmanuiacformr Mikan (P. geminatum Forsk.) from Milium 

 altemans, indicating that Milium is used in the historic sense. 



.Description. — Inflorescence of several to many dense racemes along a 

 common axis; spikelets solitary, rarely in pairs, subsessile in two rows 

 on one side of a 3-angled, sometimes narrowly winged rachis, the back of 

 the fertile lemma turned from the axis; spikelets dorsally compressed, 

 sometimes turgid; first glume usually less than half the length of the 

 spikelet ; second glume and sterile lemma equal or nearly so, 5- to 7- 

 nerved, the lemma enclosing a hyaline palea and sometimes a staminate 

 flower; fruit indurated (in the type species smooth and shining) usually 

 papillose-rugose, the margins of the lemma inrolled, the apex rarely 

 mucronate pointed. Annual or perennial, branching herbs with linear 

 leaves, the culms often decumbent and rooting at the lower nodes, con- 

 fined to the warmer temperate and tropical regions of both hemispheres. 



Brachiaria is here distinguished from allied genera by the strictly 



racemose inflorescence and reversed position of the spikelets (in which 



the first glume is present) taken in combination. 

 » 



•Schlechtendars stricture (Linnaea 26 : 537. 1853) of Steudel for including under 

 Panicum section Brachiaria Trin. a different aggregation of species from that included 

 under it by Trinius himself, must have been made in ignorance of Trinius' later paper, 

 for Steudel, not only includes the same species as did Trinius, but gives them in the 

 same order, only interspersing a few of what he considers allied species. 



