102 ANNALS OF THE ROYAL BOTANIC GARDEN, CALCUTTA. P. filaris. 
scales, unilocular from the incomplete development of the dissepiments; style 
very short; stigmas thick, elonzate-trizonous, spreading; ovules 3, basilar, erect. 
Fruit small. when thoroughly mature only slightly longer than broad. 
10-12 mm. in diameter; before perfect maturity of course smaller and very 
broadly ovoid or subobovoid; it is always rounded at both ends, but 
especially at the apex, where it is not mucronate; scales in 12-13 longi- 
tudinal series, very few in each series (5-6 only well formed), shiny, straw- 
coloured, relatively large, rather yconvex, channelled along the middle with a 
narrow dark-brown or blackish marginal line, their apices very obtuse, the margins 
finely erosely-toothed. The pericarp is very thin and’ fragile. Seed ellipsoid, 
somewhat flattened, 7-8 mm. long, 5-6 mm. broad and 3°5-4 mm. thick; its 
surfaco unequal from small depressions and straight ridges (when it is cleared of 
the crustaceous and brittle but once fleshy integument), which radiate especially 
from the chalazal fovea; the latter deep, circular, placed in the centre of the 
raphal side; albumen equable; embryo in the centre of the face opposite to the 
chalazal fovea. Fruiting perianth not accrescent, explanate, 5 mm. in diameter; 
calyx splitting down to the base into 3 very broad, almost orbicular, very 
broadly-striately veined lobes; the segments of the corolla as long as the calyx, 
ovate, and also striately veined. It is one of the tallest of Palms. 
HanrrAr.—Celebes; the Moluccas; New Guinea; Indo-China. It seems a 
Palm more frequent in Celebes than elsewhere. From North Celebes I have seen 
specimens collected by Warburg at Bojon; by the Brothers Sarasin at Tomohon in 
the Prov. of Minahassa (No. 801 in Berlin Herb.) and Koorders (No. 18427B in 
Herb. Bogdf.). ; According to Rumph it grows also in Ceram and Buru. In N.- 
W. New Guinea, Pigafetta is apparently a rather common plant. I collected 
specimens of the male plant (P. papuana Becc.) at Andai at the foot of Mt. 
Arfak, and Te/jsmann gathered the fruits in Pulo Roon (Herb. Bogor.) and Miss. 
L. S. Gibbs at Manokoari. . 4 
Its presence in Indo-China rests on some loose fruits collected by O. Kunze 
in Cambodia. | 
The rather wide geographical distribution of this Palm is probably due to its 
small innumerable fruits, provided with a rather ‘scanty but fleshy pulp, which 
‘are almost certainly used by birds for food, especially wild pigeons. which are 
some of the most effective agents of dissemination in the Papuan and Malayan 
Islands. 
OpsERVATIONS.—I have based the description above mainly on specimens 
gathered from plants cultivated in the Botanical Garden at Buitenzorg, but the 
fruits of the wild plants from Celebes and New Guinea are quite identical with 
those of the cultivated. 
I think there is only one species of Pigafetta, as the specifical differences 
between Metroxylon filare and M. elatum indicated. by Blume, Martius and Miquel 
seem to me very obscure, and I have not noticed any diagnostie character among 
the numerous specimens of Pigafetta examined by me. The size and form of its 
3 
