628 MONOECiA POLYANDRIA. Suguerus. 



the female flowers, in which case they are small, with less 

 fertile anthers, and placed one on each side of a female flow- 

 er. Calyx three-leaved ; leaflets roundish, fleshy, concave, 

 with their margins thin and imbricated. Corol of the large 

 or proper male flowers, three-petalled. Petals sub-linear, 

 smooth, expanding, of a firm leathery substance, concave and 

 yellow on the inside, and of a deep purple on the outside ; 

 margins thick and meeting only. Stamens numerous, the 

 length of the petals. Filaments short, inserted into a fleshy 

 receptacle in the centre of the flower. Anthers linear. Fe- 

 male FLOWERS sometimes in the same spadix, with the male, 

 and then just half as numerous as the male, but in general 

 they occupy separate spadices above those of the perfect 

 male, or have the less perfect male flowers which accompany 

 them, very small, probably abortive. Calyx five-leaved, 

 leaflets short, unequal, imbricated, scarcely a third of the 

 leno-th of the corol. Coro/ three-petalled, the petals cordate, 

 and much shorter than in the male. Stamejis, rather necta- 

 rial flaments, sometimes there are three, very small round 

 the base of the germ, as in Caryota ureus, sometimes they are 

 entirely wanting. Germ superior, three-lobed, smooth. Style 

 none. Stigmas three, conical and three-sided. Berry as large 

 as a crab apple, three-lobed, three-celled, smooth, fleshy, 

 when ripe yellow, pulp very acrid. Seeds one in each cell, 

 oblong, somewhat three-sided, aflixed by the lower pointed 

 end to a central receptacle, covered with a hard, black, toler- 

 ably smooth, thin shell. The embryo ot the future plant is 

 lodoed about the middle of the interior convex of the peris- 

 perm. 



The tree is nearly allied to Caryota. The chief difference 

 is in the pericarpium, which is in this a three-seeded berry, 

 in that one or two-seeded only. With respect to the various 

 and important uses of this most elegant palm I have nothing 

 to offer myself, but refer to what Rumphius and Marsden 

 have written on the subject. At the same time, I cannot avoid 

 recommending to every one who possesses lands, particularly 



