Materials for a Flora of the Malayan Peninsula. 165 



•2 to "4 in. Male jl'wers '25 in. in diara., in umbels of 3 to 8, on the 

 apices of the br.anches, or from the axils of leaves or of fallen leaves ; 

 pedicels "25 to "5 in., slender. Sepals orbicular, fleshy, concave. Petals 

 longer than the sepals, oblong, obf.use, concave. Stamens about 20, 

 forming a tetragonal mass inserted on a convex receptacle, the filaments 

 very short; anthers broad, cuneate with flat tops, 4-celIed with vertical 

 dehiscence, the connective thick; pistil 0. Female flowers apetfilous, 

 solitary, or in clusters of 2 to 5, axillary ; ovary ovoid, 5- or 7-celled ; 

 stigma large, convex with a central smooth depression, bearing many 

 black papilUe, and obscurely 5- to 7-lobed ; staminodes about 8 to 10, 

 not branclied, their heads flat. Fruit ovoid-globose, '5 to "75 in. in diani., 

 orange-coloured, pulpy, with a thick fleshy apiculous crowned by tlie 

 persistent stigma. Hook. hi. Fl. Br. Ind. I, 263. Pierre Floie Forest. 

 Coch.-Chiae, fasc. VI, p. xxix, (excl. t. 81. fig. F.) G. parvifolia, ]\Iiq. 

 Ann. Mus. Lugd. Bat. I, 208. Rhinostigma parvifolium, Miq. Fl. Ind. 

 Bat. Supp. 495. 



It) all the provinces ; in ti'opical forests. Distrib. Sumatra. 



This is one of the commonest species of the genus. The pulpy 

 fruit is eaten by the aborigines. Griff. No. 854 and Maingay Nos. 152 

 and 162 are the specimens on which Planchon founded the species. Griff. 

 Nos. 865 and 867 (referred by Planclion and Triana and also by Piei-re 

 to G. Kydiana, Roxb.) in my opinion fall here, as also does G. timbellifera, 

 Wall. Cat. 4864 ; but Anderson reduces the latter to G. Omva, Linn. Pierre's 

 figure, (t. 81, fig. F.), which he names S-nigro-lineata, does not represent 

 the flowers of the type specimens in the Calcutta Herbarium which bear 

 the numbers which Pierre quotes. I fear therefore that there must 

 have been some confusion in the distribution of the Griffithian collec- 

 tions. 



It is quite possible that the description which I have given above 

 may cover two species. The specimens with lanceolate-acuminate leaves 

 have rather more erect and fainter nerves than these with ovate-lanceolate 

 caudate-acuminate leaves. But, although I have dissected a large nnrn- 

 bei' of the male flowers of each, I cannot detect any tangible difference. 

 Unfortunately I have been able to find very few female flowers. An 

 examination of Miquel's type specimen of his Bhinostiqma jhirvifoliuin 

 leaves no doubt whatever that it is identical with Pianchon's G. nigro- 

 lineata. 



I never find the petals reflexed ; but Ander.son, in Hook. fil. PI, 

 Br. Ind. ( 1. c. ), describes them, and Pierre (1. c), figures them, as reflexed 

 from about the middle. 



26. Garcinia KuNSTLEtti, King, n. sp. A shrub, 6 to 8 (rarely 15) 

 feet high ; the young branches dark-coloured, not-angled. Leaves mem- 



105 



