188 liOlIINSON. 
J 
duplicate is preser\cd at Kcw, and a fragment from this is in tlie her- 
barium of the Bureau of Science, well matched by another collection from 
near the type locality. But nothing in this herbarium sufficiently repre- 
sents Xaves' plate, which may well have been inaccurate; its closest match 
is a specimen which can not at present be distinguished from a species 
described from Singapore, but unfortunately published after C. hlancoi 
Eolfe, which is here regarded as a doubtful species. Eolfe's species 
antedates VidaFs and following the Vienna code, the name G. hlancoi 
could be shifted about from one species to ihe other, according to varying 
opinion as to the validity of the older. ■ The writer, feels certain that 
the best plan here, as in all similar cases, is to reject the later name, 
and believes that no ambiguity will ever be possible if Vidal's species, 
which is certainly not that of either Blanco or Bolfe, is hereafter known 
by the name of the botanist who described it: it is accordingly here 
renamed C. vidalii. 
KEY ^ TO THE PHILIPPINE SPECIKS. 
F 
Styles of the female flower 2-cleft. 
Ovary and capsule more or less S-lobed. 
Ovary glsibrous. 
Leaves less tliati 5 cm long - 1. C. or<jyali» 
Leaves at least 8 cm long. 
Pairs of lateral veins 9-15. 
Leaves 4 times as long as wide _... 2. G. apiculatus 
Leaves about 3 times as long as wide 3. C viyrianthus 
Pairs of lateral veins over 20 4. C. cupreus 
Ovary more or less pubescent. 
Leaves slenderly acuminate. 
Pairs of lateral veins less than 10. 
Venation prominent. 
Capsules pubescent 5. 0. hridclifolius 
Capsules subglabrous. 
Capsules very conspicuously reticulate 6.(7. venosus 
Capsules not conspicuously reticulate 7. (7. laevis 
Venation obscure 8. G. vidalii 
Pairs of lutoral veins mure thiiu 10. 
Leaves abruptly acuminate 0. G. cverettii 
Leaves gradually acuminate 4. G. cupreus 
T^eaves shortly and broadly aciiniinate. 
Leaves ovate „ 10. G. ovatus 
Leaves elliptic or oblong 11. G. decipicns 
Ovary and capsules not 3-lobed 12. 0. inleyer 
Styles of the female flower 4-cleft 13.(7. (luadrifidus 
* 
* For the purposes of this key it has been assumed that when the capsules 
show any traces of pubescence that the ovaries have been pubescent. The ovaries 
of G. cuiireus being unknown, and the capsules being glabrous, it has been 
included under both divisions of the key. In the text the disk of the female 
flowers lias been described as it was foimd. It is probable that the apparent 
lobes are the result of mechanical injury, caused by the expansion of the growing 
capsule. 
