304 MR. T. F. CHEESEMAN ON 
ore nud ; ligula membranacea, brevissima, truncata, denticulata ; laminze anguste 
lineares, plan vel siccitate subeonvolutz, 12-16 em. long:e, 2:5 mm. latee, sensim 
tenuissime acuminate, erecto-patule, preeter marginem scabrum lzeves, tenuinerves. 
Panicula lineari-oblonga circ. 9 cm. longa, stricta, haud vel vix basi interrupta, 
rhachi levi, ramis binis vel ternis 1:5-2:5 em. longis suberectis simplicibus a basi 
spiculis densiuscule obsitis. Spiculze secus paniculse ramos gemine, altera pedicello 
brevissimo, altera eam subequante crassiusculo subclavato fulta, lineari-lanceolate, 
3:5 mm. long. virides. Glume steriles subzequales, lineari-lanceolate, acute, ex 
apice obsolete bidentulo mucronem 0:5-1 mm. longum, scabrum, exserentes, 3-nerves, 
nervis valde prominentibus superne aculeolato-scabris, inferior basi (in callo) minute 
barbulata. Gluma florifera sterilibus subbrevior, lanceolata, tenuissime 3-nervis, 
glaberrima, inter apices dentium brevissimorum aristam emittens tenuem haud geni- 
culatam neque tortam sed leviter flexuosam 5 mm. longam (giuma ipsa fere duplo 
longiorem) scabram. Palea gluma paullo brevior, lanceolata, acuta, binervis. 
Caryopsis lanceolato-fusiformis, 1:5 mm. longis, embryone dimidiam caryopsin 
eequante, macula hilari punetiformis. 
* Rarotonga in saxosis montium rara. Legit T. F. Cheeseman, Jul. 1899. 
* Affinis G. stricte, Brongn., que vero abunde differt foliis late linearibus ore 
pilosis, paniculee interruptee ramis verticillatis fastigiatis, spiculæ gluma inferior basi 
distincte barbata (barba circ. 5 glum:e :equante) gluma florifera (saltem in planta typica) 
mutica. G. Fergusonii, Trim., habitu nostre similis, differt foliis pilosis, spiculis 
majoribus (5 mm. longis), gluma inferiore mutica." 
I had referred this to G. stricta, but Prof. Hackel states that it is clearly distinct, and 
has supplied me with the above diagnosis and notes. 
MISCANTHUS JAPONICUS, Anderss.?— Native name Kakao. Very plentiful on open 
hill-sides throughout the island, and much used for lining the native houses, Ae, My 
specimens are long past flower, and only show old panicles from which the spikelets 
have fallen, so that some little doubt attaches to the determination. 
Pacific Islands northwards to Malaya, the Philippines, China, and India. 
ER ed E du D "s S D 
IANTHUS MAXIMUS, Brongn. ?—Hill-sides above Arorangi, rare. I am indebted to 
Dr. Stapf for suggesting that this may be Brongniart's E. maximus, a species of which 
there is no authentic example at Kew. 
Tahiti. 
ANDROPOGON ACICULATUS, Retz.—Common in open places throughout the island, 
both near the sea and on the mountains. 
Pacific Islands, Australia, and northwards to India and China. 
ANDROPOGON REFRACTUS, R. Br.—Open rocky places on the hills, not uncommon. 
New Hebrides, Fiji, Tahiti, Australia. 
