186 
Card. Bull. Singapore 69(2) 2017 
2a. Bracts caducous before the flowers open.3 
2b. Bracts persistent at least until the fruits mature.4 
3a. Both leaves and bracts hairy with glandular hairs; inflorescence apex ± pointed, 
not fully covered by bracts; bracts narrowly triangular, c. 1.2 mm long and 0.5 mm 
wide, brown with central purple stripe. E. elongata 
3b. Both leaves and bracts glabrous; inflorescence apex ± rounded, completely 
covered in imbricate bracts; bracts ovate, 2-3 mm long and 1-1.5 mm wide, 
whitish or pinkish. E. pallida 
4a. Bracts glabrous, some of them persistent after the fruits have fallen; inflorescence 
elongate, 2-6 cm long; corolla caducous; endemic to Sabah. E. confusa 
4b. Bracts minutely ciliate, all of them shed with fruits; inflorescence shortly 
cylindrical, up to 2.5 cm long; corolla persistent; endemic to Thailand. 
. E. compressa 
5a. Each flower subtended by a bract and pair of subulate bracteoles; bracts ± patent 
during anthesis, linear with sharply upturned obtuse apex; fruit longer than wide, 
much shorter than sepals. E. papuana 
5b. Each flower subtended only by a bract, bracteoles absent; bracts ± erect during 
anthesis, with straight acute or slightly curved apex; fruit as long as wide, as long 
as sepals or longer.6 
6a. Bracts lanceolate, 2-2.5 mm long and 0.5-0.6 mm wide, brown with purple 
central stripe, usually persistent after the fruits have fallen; inflorescence apex not 
completely covered by bracts; fruit as long as sepals. E. kinabaluensis 
6b. Bracts ovate, 1.5-2 mm long and 0.8-1.2 mm wide, brown with basal purple 
patch, usually shed with fruits; inflorescence apex completely covered by 
imbricate bracts; fruit longer than sepals. E. cylindrica 
Conservation status of Epirixanthes 
The forests of Borneo, including Brunei, Sabah and Sarawak, harbour the highest 
diversity of Epirixanthes. Six species have so far been reported from this island 
(Van der Meijden, 1988; Coode et al., 1996; Tsukaya et al., 2016). Similar to other 
mycoheterotrophic plants, all species of Epirixanthes are closely tied to primary or 
very lightly logged forests and as such are threatened by continuing deforestation. 
However, published data on their distribution and ecology are very scarce and most 
of the species are poorly represented in herbaria. For these reasons, we propose that 
all species of Epirixanthes in this study be evaluated as data deficient (DD, IUCN 
Standards and Petitions Subcommittee, 2016) which emphasises the need for further 
study and sampling. 
