110 NOTE ON A CRITICAL CHINESE GRASS, 
hills around Canton, a Grass, which I subsequently distributed to 
various herbaria, with the doubtful determination of Chameraphis 
aspera, Nees ?, adducing, also with doubt, as a synonym the Pseudora- 
phis Brunoniana, Griff. Mr. Bentham, however, informs me, on the 
authority of Colonel Munro, that Griffith’s plant is C. depauperata, 
Nees, and he is inclined to think mine is C. hordeacea, R. Br. I may 
be allowed to observe, however, that the Canton plant has not at all the 
* thyrsus simplicissimus, bipollicaris, Hordeum distichum referens," as- 
signed by Steudel (Synops. Glumac. 49) to R. Brown’s species, nor 
does it agree better in inflorescence with the detailed description given 
by Kunth, from Brown’s own specimens, in the supplementary volume 
of his ‘ Agrostographia.’ On the other hand, it has a “ panicula ter- 
minalis, ovata, effusa, subglabra, inferne subexcavata, floribus 1 v. 2 
infimis sessilibus, in excavationibus nidulantibus, superioribus stipi- 
tatis, stipite ultra florem terminalem in aristam subulatam denticula- 
tam flore duplo longiorem producto denticulato scabro, denticulis 
antrorsis," as Griffith describes (Notul. ad Pl. Asiat. pt. iii. p. 29). 
That author states the upper florets to be stipitate and usually solitary, 
but in his figure (Ic. Pl. Asiat. pt. iii. t. 145, ** Panicum Brunonia- 
num") the lower and middle radioles are depicted with usually two 
distant spikelets; the upper with one only, all sessile or subsessile, 
and this is nearly always the case with the Chinese Grass, though the 
middle branches have sometimes 3—4 spikelets. The plate itself might, 
so far as accuracy is concerned, have been drawn from my specimens. 
I possess, from Dr. Thwaites, examples of C. aspera (C. P. 3846) 
and C. depauperata (C. P. 3857),—the determinations having, I doubt 
not, been verified at Kew. The latter, with its dense, spiciform, 
barley-like panicle, with short, 1-spikeleted, appressed rays, is altogether 
unlike Griffith’s plate; the former differs from the Canton Grass by à 
rather more robust habit, laxer vagine, extreme scabridity of them and 
the leaves (which, except at the margins, are smooth in my plant) less 
patent and very flexuose panicle-rays, bearing sometimes as many as 
12 approximate, usually overlapping, not distant spikelets. I cannot, 
therefore, help suspecting that Mr. Bentham’s opinion has arisen from 
some misconception, and that the Canton and Bengal plants are iden- 
tical; and if, as seems to be the case, we have here a new species, 
holding a middle station between C. aspera and C. depauperata, it 
might well bear the name of Panicum (Chameraphis) intermedium, 
