black bases and tips 10-20 X 2-3 mm, with few on rachis. Inflorescence bear- 

 ing two subequal bracts, the outer one a prophyll 25 cm long, papery and 

 lacking spines, the inner a peduncular bract, brown, 30 X 2 cm, covered with 

 many flattened black to white spines to 5 mm long; axis ca 1 5 X 7 mm with ca 1 2 

 rachillae, each ca 13 cm long. Staminate flowers yellowish. Fruits dark purple 

 when ripe, flattened-globose, 1.5 cm in diameter, with a 2 mm long stigmatic 

 residue at the apex. 



Distribution (B. maraja): Colombia, Guyana, Surinam, Peru, Brazil, 



Bolivia. 



This palm is identified only provisionally as B. maraja, owing to 

 the confused taxonomy of this poorly known genus. According 

 to Moore (pers. comm.) it appears to key out in Drude's treat- 

 ment in Flora Brasiliensis (1882) to either B. chloracantha 

 Poepp. or B. pallidispina Mart. Macbride's description (1960) of 

 B. chloracantha notes a stem ". . .8 cm in diameter, more or less 

 aculeate with straw-colored, subterete spines. . . " and leaf seg- 

 ments with setulose margins. The palm in question, as noted in 

 the above description, has a much smaller stem diameter, both 

 light- and dark-colored, flattened spines, and smooth leaf mar- 

 gins. Wessels Boer's (1965) description of B. pallidispina (which 

 he reduced to synonomy under B. maraja) fits this palm, except 

 again for the ciliate margins of the pinnae. Martius (1847) pub- 

 lished a plate of B. pallidispina along with descriptions of that 

 species and of B. maraja. From these descriptions, my material 

 from "Las Gaviotas" agrees very closely with B. maraja. The 

 type specimen, Poeppig 2107, appears of little value in clarifying 

 the determination, as the photo shows it to consist of young 

 leaves and an immature fruiting panicle. However, from the 

 photo it can be seen that the perianth of B. chloracantha, which 

 persists in fruit, is mucronate, in this apparently differing from 

 the "Las Gaviotas" material which lacks the sharp point. I am 

 unsure if this structure is stable or variable in the bactroid 



alliance. 



Wallace (1853) described fruit of B. maraja as having ". . .a 



thin pulp of an agreeable sub-acid flavour — a pecularity not 



found in the fruit of any other American palm that I am 



acquainted with." This taste is characteristic of the palm in the 



Llanos. However, according to Moore (pers. comm.), a subacid 



flavor is probably characteristic of fruits of several species of 



8 



