170 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 13, NO. 9 
Trinidad; Lindman A 2597, and Glaziou 14397, Brazil; Hassler 12444, 
13021, and Rojas 3071, Paraguay; and Ekman 675, Argentina. The 
blades of the Raddi specimen are shorter in proportion to their width 
than common, as in Hassler 13021. In Grasses of the West Indies® 
the key is misleading in that the fruit of Olyra ciliatifolia is said to be 
“clothed with thick silky hairs at base and summit.’ The fruit is 
loosely pubescent throughout. 
32. OLYRA FLORIBUNDA. “Ad radicem Montis Corcovado.” The 
specimen consists of a tuft of five flowering culms and two sterile ones. 
Raddi cites ““Raddia brasiliensis. Bert. Opuse. Scientif. di Bologna 
1819. T. III p. 40.” Bertoloni’s detailed description agrees in every 
way with Raddi’s specimen and was undoubtedly drawn up from a 
duplicate. (There seems to have been abundant material of this collec- 
tion; there were two full sheets of it among the unidentified grasses in 
the herbarium at Florence, as well as a specimen in the Delessert 
Herbarium.) The distinguishing generic characters given are the 
distinct staminate and pistillate inflorescences. Bertoloni states that 
the generic name is given as a just tribute to Raddi and that the 
specific name commemorates his courageous voyage to Brazil. But 
Raddi declines the honor conferred on him and renames the species 
Olyra floribunda. He describes the distinct staminate and pistillate 
inflorescences but states that the style and stigmas are as in the three 
preceding species, and that they agree with the generic characters 
assigned by Swartz to Olyra. We recognize Raddia as a distinct 
genus, and this species as R. brasiliensis Bertol. The valid name 
under Olyra is O. brasiliensis (Bertol.) Spreng. I have seen no other 
collection which agrees exactly with the Raddi specimens. Glaziou 
4336 and 12265, Brazil, probably belong to this species, but the blades 
are more than twice as large and the pistillate spikelets are larger and 
have a longer acuminate tip, but the spikelets are clothed, as in the 
Raddi specimens, with a dense short retrorse pubescence, with a few 
long stiff hairs intermixed. 
33. PHARUS BRASILIENSIS. “Prope Rio-Janeiro.’”’ The specimen 
is a complete plant with a very immature panicle. The name on the 
folio is not “brasiliensis” but one that was not published. The plant 
agrees with Raddi’s description, and as but one species of Pharus is 
given there can be no doubt that this is the type of P. brasiliensis. 
It agrees with rather narrow-leaved specimens of Pharus glaber 
H. B. K., such as Blanchet 1018, Mosén 1778, and Dusén 91, from 
* Hitchcock & Chase, Contr. U. 8S. Nat. Herb. 18: 357. 1917. 
