72 
MARTELLI. 
have the leaves long acuminate and agree perfectly with Elmer’s type specimens 
no. 7201 ; on the contrary the leaves of some other specimens are shortly 
acuminate. The marginal teeth of the leaves also seem rather variable as to 
size and shape. Some of my specimens have the peduncular part of the raceme 
1 m in length, and the syncarps, not very densely grouped at the summit of the 
peduncle, are 9— 1 4 cm in length, always subtrigonous but varying from cylindrical 
to obovate. Pandanus Gopelandii Merrill seems to have a rather wide distribution 
in the Philippines and it is therefore probable that some of its specific character- 
istics are not absolute and that later we may be able to distinguish various local 
forms of it. 
SPECIES DUBIAE. 
1. Pandanus Sabotan Blanco FI. Filip. (1837) 779. 
See above under Pandanus tectorius var. sinensis. 
2. Pandanus inekmis Blanco FI. Filip, ed. 2 (1845) 537. 4 
Blanco’s description of this species is so short that it is impossible for me 
even to surmise to what kind of plant it may be applied, it is doubtful even 
if it is a Pandanus because of its: “Hojas esparc-idas,” but if it is a Pandanus 
I do not know any unarmed except Pandanus tectorius var. laevis Warb. ( Panda- 
nus laevis Kunth), of which only the male plant is known. This variety, 1 
think, does not grow spontaneously anywhere, but is a domestic form, multiplied 
by agamic reproduction, widely distributed and cultivated in many places, its 
pollen being used as a hair powder. 
3. Pandanus latifolius Perr. in Mem. Soc. Linn. Paris, 3 (1824) 134. 
From a small island at the entrance of Basilan Strait, imperfectly described. 
(See Pandanus dubius Spreng, p. 67 above.) 
4 Pandanus inermis Blanco is almost certainly the same as Dracaena angusti- 
folia Boxb., a common and widely distributed species in the Philippines, and 
which Blanco does not consider in his “Flora de Filipinos” unless as Pandanus 
inermis. (E. U. M.) 
