Polygala papuana
Content
Description
Liana, glabrous in all parts except sepals and seed.
Leaves obovate-oblong to obovate-lanceolate, rarely lanceolate, c. 3.5-8.5 by 1.5-3 cm, shortly acuminate, gradually attenuate at base, nerves c. 5-7 pairs, 0.5-1.5 cm petioled.
Flowers 4.5-6.5 mm long, white with red keel, on 3-5 mm long pedicels.
Sepals adnate over 0.5-1 mm to the petals, ± orbicular shortly ciliate, the alae broadly obovate.
Ovary obovate, apically with a 1 mm long sterile part.
Capsule obovate-linear, c. 18-22 by 2-2.5 mm, shortly acuminate at apex, gradually narrowed towards base, unwinged.
Seed 5-1 by c. 1 mm, completely covered by white silky hairs c. 1.5 cm long.
Notes
Recent Australian authors () claimed that this species belongs to the genus Comesperma. They agree with VAN STEENIS (l.c.) that Comesperma differs from the South American genus Bredemeyera in a single character; in the latter the carinal appendage bears the coma-hairs; in Comesperma the coma-hairs are present all over the seed or in two rows. It is dubious whether other differences exist. However, it is uncertain whether Comesperma itself is a good genus. Some Comesperma species (viz. those of sect. Prosthemosperma F.v.M. Pl. Vict. 1862, 186) have no coma-hairs and no elongated capsule. Comesperma can only be upheld if it could be proven that this is a reversion (by reduction) of the evolution. If, however, these 'abnormal' Comesperma species represent the primitive character of fruit and seed, the genus cannot be upheld against Polygala. Thusfar this important question has been neglected. On the other hand it should be noted that two species of Polygala sect. Chamaebuxus (P. wattersii HANCE and P. mariesii HEMSLEY, both from China) have 'typical' Comesperma fruits and seeds, but definitely do not belong to this group as can be demonstrated by the differences in floral characters. Thus there is good reason to assume that Comesperma should be merged into Polygala. The same may be true for Bredemeyera, but the present state of knowledge on this genus is still incomplete. In view of the weak differences of Bredemeyera and Comesperma with Polygala, there is no good reason to combine both (under Bredemeyera) as VAN STEENIS (l.c.) proposed. As a consequence Bredemeyera papuana is best considered for the present to be a species of Polygala. The affinity of P. papuana with the typical representatives of Comesperma is unclear. Surprisingly, PEDLEY (l.c.) dismissed the difference in the calyx as 'rather trivial', whereas CHODAT used this character as the main one for his subdivision of the genus Polygala. As far as I am aware only one other species (P. acicularis OLIV. from tropical Africa) has caducous alae and persistent sepals. At present it seems better to keep P. papuana in a separate section, separate from Comesperma when considered as a section from Polygala.