Iriartea.'] equatobial-amebicak palms. 133 



thus assumed to be generic Lave their analogues In genera Of 

 exogens, where they are not considered of equal importance. 

 Take, for example, the genus Cordia, which has an ovary of four 

 uniovulate carpels, followed commonly by a monocarpellary fruit, 

 tipped at the apex with the persistent stigmas, or at least with a 

 stigmatic scar. But in the Cordias of the " nodosa " group (and 

 I refer especially to one of my own gathering, C. umhraculifercs 

 var ? PL Exsicc. 123-i) the ripe fruit is a small, yellow, obliquely 

 oblong, and mostly 1 -seeded drupe, which increases in gibbosity 

 as it ripens, so that when quite ripe the stigmatic scar is midway 

 between the base and the geometrical apex, and the persistent 

 calyx looks as if grown on to the side. But when two or more of 

 the carpels are fertile, the obliquity usually disappears ; while 

 many species of the genus have even solitary ripened carpels sym- 

 metrical. 



I was able to examine fully but two species oi Iriartea^ whose 

 descriptions I reproduce below ; but I fell in with some others, 

 and /. exorrhiza^ especially, is one of the commonest of Amazon 

 palms, extending through the whole length of the great plain and 

 some way up the moist Andine valleys, and northward as far 

 as the borders of the savannahs of the Orinoco. In the Peruvian 

 Andes, at from 2000 to 4000 feet, /. delioidea grows along with it ; 

 and in the Equatorial Andes, at the same elevation, I have seen a 

 very fine large-leaved Iriartea which I suppose undescribed. 



1. I. VENTRICOSA, Mart, Palm. 37; t. 35^ 36 : caudice procero 

 medio ventricoso; foliis 18-pedaIibus pinnatis glabratis, pinnis pluri- 

 jugis magnis cuneato-flabellatis apice sinuato-pra3morsis demum 10- 

 partitis ; spadicibus vix bipedahbus^ ramia 3-pedalibu3 pendulis; 

 spathis 10-12 ; baccis globosis nigrescentibus. 



into not more than ten filaments, which I find slightly anastomosed, although 

 Wendland describes them as quite free. 



In/, ventricosa, Mart. {Deckeria, Karst. ; Iriartea, W end!.), the embryo is 

 lateral a little below the middle of the seed, and the filaments of the raphe di- 

 stinctly anastomose. 



Braz 



I 



them the same generic name (Faxiuba); and a closer examination reveals so many 

 essential resemblances, that the botanist will probably indorse the opinion of the 

 savage, and continue to unite these palms under one title. 



CeroxyUm 



roots 



from Iriartea, 



