Hyparrhenia.\ clvii. gramine^e (Stapf). 
341 
bearded. Glumes equal, subchartaceous, lower narrowly truncate, 
flattened on the back and slightly depressed along the middle nerve, 
rounded on the sides with involute margins to beyond two-thirds, then 
finely keeled with the keels rigidly ciliolate, very faintly 9-nerved, more 
or less pubescent with white hairs ; upper deltoid-truncate or obscurely 
mucronulate, 3-nerved, ciliolate upwards. Lower floret reduced 
to a hyaline linear-oblong truncate finely 2-nerved sparingly ciliate 
valve. Upper floret ^ : valve stipitiform, 2-dentate, teeth almost 
subulate, glabrous ; awn slender, 4-5 lin. long, column dark brown, 
minutely fulvously hirtellous ; valvule 0. Anthers almost 1 lin. long. 
Pedicelled spikelets neuter, linear-lanceolate, 2J-3Jlin. long, reddish ; 
lower glnme thin, subherbaceons, 9-11 -nerved, acute, usually mucro- 
nate or shortly aristulate, hairy like the fertile spikelets, rigidly 
ciliolate along the keels, one side sometimes slightly involute ; 
upper equalling the lower, very acute, 3-nerved, ciliate upwards ; 
lower valve oblong, subobtuse, slightly shorter than the glumes, 
1-nerved, ciliate ; upper very narrow, usually much shorter and 
nerveless. Homogamous spikelets very similar to the pedicelled 
member of the heterogamous pairs, but rather larger, broader and 
flatter —Andropogon formosus, Hort. (partly ?). 
Nile Land. Abyssinia : Tigre ; without precise locality, Schimper, 1009 ! 
Hackel attributes the name Andropogon formosus to “ Klotsch in h. berol,” 
and adds “et in hort. mult.” According to Jaeger (J) in Gartenfl. xi. 241, 
a grass named Andropogon formosus had, at the date of his writing (1862), been 
for a long time in cultivation in the Botanic Garden at Berlin, and he himself 
received it from there in 1856. Pynaert (in FI. Serr. 2 me ser., iv., 1859, 56) gives 
a figure of it and some cultural notes, taken from Neubert, Deutsch. Mag. /. 
Garten- u. Blumenkunde about 1858. It became more generally known after 
1860 and is mentioned repeatedly in horticultural literature ; but it was never 
described, whilst the figure reproduced by Pynaert and others allows of no 
identification. It is possible that the name came in time to cover more than 
one species. However, specimens grown at Kew from 1878-1882 agree with 
Schimper’s plant quoted above, and it is very probable that the species was 
raised at Berlin, like many others, from seed communicated by Schimper, though 
not from no. 1009, which belongs to the set distributed in the sixties as “ Plantse 
Abyssinicse ex Tigre v. Begemder. Collegit Schimper, a. 1863-8.” But all the 
numbers of that set below 1100, so far as I have been able to check them from 
the original tickets, were actually collected in Tigre in 1862. 
29. H. Schimperi, Anderss. in Schweinf. Beitr. FI. Aethiop. 306, 
Perennial, up to over 6 ft. bigb, with extra vaginal innovations, 
tbeir catapbylls short, ovate, glabrous. Culms geniculately ascend- 
ing, terete, glabrous, smooth, slender and wiry, up to 2J lin. thick in 
strong specimens, 7-8-noded below the panicle, with mixed branches 
from some or most of the nodes. Leaf -sheaths terete, glabrous, 
the lowest withering away, the middle and upper usually much 
shorter than the internodes ; ligules scarious, rounded or truncate, 
about 1 lin. long ; blades linear from an almost equally wide or more 
or less attenuated base, long-tapering to a fine point, up to over 1 J ft. 
long by 2J-4J lin. wide, flaccid, pale green, sometimes tinged with 
