213 
data vel obtusa, vel si subacuta nunquam acuminata, basi cuneata vel 
interdum fere rotundata, subtus pallidiora, venis primariis lateralibus 
utrinque circiter 6-10 subtus sat stent stipulæ minutissimæ vel 
obsolet s Cyme terminales, pauciramose, breviter pedunculate, quam 
o breviores, ebracteate, shiite ola tæ. Flores albi (Comins), 
anik fere glabri, vix 3 as longi, calycis dentes minuti, ovati, 
obtusi; corollz lobi oblongi, obi: tubum subzquantes, fauce longe 
villoso-barbata; stamina styloque breviter exserta. Capsula circiter 
3 lineas lata, glabra 
This was also collected by the Rev. R. B. Comins, who localised it as 
found on rocks by the seashore. 
COMPOSITÆ. 
Blumea Balfourii, Hemsl., n. sp. 
ilnei similis sed aT paucioribus subduplo majoribus, 
involueri bracteis numerosioribus exterioribus latioribus crassioribus, 
acheniis distincte striatis minus hirsutis, pappo albido. 
It is very difficult to define the limits of the species of Blumea, and 
the investigation of living plants might xu to combining this with 
D. Milnei, Seem. (Bot. Vo oy. * Hers ald,” 141, t. 27), for they are 
oe much alike in foliage. ‘The name of oni A. F. Balfour, 
„is associated with this plant in recognition of the facilities he 
aforded for making a ‘bolanieal collection in the Solomon Islands. 
ASCLEPIADE. 
Hoya inconspicua, Hemsl., 
oy Sp. 
Species foliis anguste TAMEN floribus inter minores propria. 
Scandens, parasitica, undique glabra, ramis graciliusculis. Folia 
primaria erorum sepe a Rs tos radiciformes 
folia ramulorum  secundorum valde abbreviatorum ex 
axillis (?) filamentorum erumpentia seepius ina r 
eviora, utriuque e a en 
petiolus circiter 3 lineas longus, sæpe tortus. LAE M pauciflorz, 
pedunculatze, pedicellis iet 4—6 lineas longis. Flores circiter 
-5 lineas diametro; calycis segmenta minuta, prati obtusa ; corolla 
lobi ovati, acuti, apice recurvi, intus, ut videtur, rugulosi; c orone 
segmenata inflata sursum inata deorsum bifida vel tanis 
bi ^ 
n 
wan pair of leaves ; but the exact nature of their a cannot 
be determined peste the imperfect specimens we possess 
CYRTANDRE.E. 
Boea Commersonii, R. Br.—This is oe from several localities 
in the New Britain Arthipelago, from North Australia, and from Java ; 
but Kew E deeem only one vies tm and that is also from 
the Solomon Islan 
Epithema, s species PORCH ets there is no doubt concerning the 
affinity of this plant, the flowers are too rotten for description. 
u 82451, B 
