— 139 ^ 



very exact description of that species disagrees with the present by the 

 following important points: 



"Folia petiolata, ovata. Bracteae rotundatae. Bracteola calyce aequilonga 

 tubo corollae duplo brevior. Labellum ovatum acutum, planiuscuium, basi 

 utrinque bilobum, maculis et lineis sanguineis pictum. Filanientum coccineum. 

 corniculum purpureum." To me it seems impossible that Koening could 

 have meant Zingiber Casuinunar by this description. Regarding the shortness 

 of the bracteola it must be related to Z. odoriferum 81. and Z. acuminatum 

 Val. In all other species known to me the bracteola equals the corollatube, 

 at least to 2/3 of its length. 



9. Zingiber gramineum 81. 1827, 45; Qagnepain 1908, 81; — Z. 

 datum 81. 1. c? K. Schum. 1904, 175 ; — Z. (an Roxb?) Val 1804, 7 ; 



— Z. alliaceum K. Sch ! l.c, (non [Donacodes alliacea Tet B, quae est 

 Homstedtia alliacea Val.), lab. nostra. XXIll. 



The two forms described by Blume as Z. gramineum and Z. elaium 

 are probably not specifically different as already suggested by K. Sen. 

 Whether they really are identical with the Roxburghian spec, as 1 accepted 

 in my former paper (and 1 think it still most probable) is not to be 

 decided with absolute certitude. But the plant depicted by Roscoe as 

 Z. elatum, considered by Baker as a variety of Z. capitatuni Roxb does 

 certainly not resemble Z. gramineum 81. 



Hasskarl (1843, 122) considered Z. elatum 81. and Roxb. as a top- 

 flowering variety of Z. Cassumunar, a modification he says, often occur- 

 ring, and does not mentiot» Z. gramineum 81. But 1 saw in Leyden a speci- 

 men of the bot, gard. of Calcutta (with a lateral scape) named Z. elatum 

 and this resembled very much Z. gramineum 81. This specimen was 

 determined by K. Sen. as Z. alliaceum K. Sch. and agrees with his scanty 

 description of that species and of Z gramineum 81. 



Qagnepain (1908, 81) gives a good description of Z. gramineum, but 

 he did not see terminal inflorescences, which are not uncommon in the 

 Java plants if grown from seeds, nor did he know the flower. I therefore shall 

 add a few details and a drawing of this species: 



Inflorescence long pedunculate now terminal on a foliate stem, now 

 lateral on a sheathed scape, young obovate and enclosed by the two large 

 oppressed empty bracts, accrescent and becoming broadly ovate, blunt. 

 Bracts elliptic (30X14) Of elliptic lanceolate thinly pilose with a much 

 narrowed Curved penicillate tip, loosely imbricate with free patent tips. 

 Bracteoles as long as the bracts (28X10 mm.) and similarly shaped but 

 narrowed and with an acute not itcuminate top, pilose. Flowers small, 

 white, 35 ntni. long. Labellum round ( 15 — 15 X 17) quite entire, pale yellowish. 

 Stan>inode8 obsolete or very small, elliptical, white, facing the labellum 



