HITCHCOCK GRASSES OF CUBA. 239 



Upper floret trimcate-dilated. 



Awns 1 to 2 mm. long; lower lemma long-ciliate on 



the keel and lateral veins, but not at apex 1. C. cUiata. 



Awns, or some of them, 5 mm. long or more; lower 

 lemma ciliate on the upper part of marginal nerves, 



not on keel 5. C. paraguaiensw. 



Upper floret narrowed toward apex. 



Lower lemma strongly ciliate at apex with tuff of 



hairs 2 mm. long 3. < . i legans. 



Lower lemma only pubescent at apex. 



Lower lemma 3 mm. long; blades short and flat. 



abruptly rounded at apex 7 . C. radiata. 



Lower lemma 2 mm. long; blades mostly invo- 

 lute-pointed 4. C. eh usinoides. 



1. Chloris ciliata Sw. Prod. 25. 1788. 



Trinidad, May 17. Wright 743; Vento, Baker HC 1184, Curtiss (500; Guanabacoa, 

 Leon 185 in part; Santiago de las Vegas, Tracy 9115, Hitchcock in 1906; Triscornia 

 9085; Habana, Tracy 9106; Herradura, Hitchcock in 1906: Arroyo Apolo, Leon 574. 



2. Chloris cruciata (L.) Sw. Prod. 25. 1788. 

 Agrostis cruciata L. Syst. Xat. ed. 10. 2: 873. 1759. 



Chloris brevigluma Wright, Anal. Acad. Cienc. Habana 8: 200. 1871; Sauv. Fl. 

 Cub. 191. 



Bushy savannas, Hanabana, May 16, Wright 1549; Punta Brava, Baker HC 4067; 

 Guanabacoa, Baker HC 2927, Curtiss 584, Hitchcock in 1906; Madruga, Britton & 

 Shafer 604 in Herb. X. Y. Bot. Gard. 



There are three Wright specimens of this species in the Grisebach Herbarium: 

 Nob. '917=1548" and '-932=1548" from western Cuba, 1863, and no. 1549 from 

 eastern Cuba, 1859. The type of C. brevigluma is in the Gray Herbarium with 

 printed label for 1860-64, no. 1548. In Sauvalle's Flora Cubana the type is mis- 

 printed "1848 p. p." Wright seems to have distinguished his species from C. eleu- 

 sinoides mounted on the same sheet (Wright 1549), which he regarded as the true 

 I ". cruciata. Wright's type matches his no. 1549 in the National Herbarium. 



3. Chloris elegans 11. B. K. Nov. Gen. & Sp. 1: 166. 1816. 



Vento, Shafer 483, Baker HC 1183; Madruga, Britton & Shafer 725: Mazarra, Baker 

 HC 4023. 



The type of this has not been examined, but the specimens cited above agree 

 well with the plate accompanying the original description . a 



4. Chloris eleusinoides Griseb. Fl. Brit. W. Ind. 539. 1864. 



Chloris eleusinoides vestita Greenman in Combs, Trans. Acad. St. Louis 7: 477. 1897. 



Sandy pine woods, La Griza, January, Wright 3819; Wright 3818; Wright 1548; 

 Baker HC 4067; La Magdalena, Baker PL Trop. Am. 4; Havana, Leon 287; Herradura, 

 Hitchcock in 1906; Cienfuegos, Combs 631 in Gray Herbarium. 



The Grisebach specimen is from eastern Cuba, 1859, no. 1548. Xos. 1548 and 

 1549, in the Engelmann Herbarium, both from eastern Cuba. 1860, are Chloris 

 i h usinoides. Ko. 1549 in the Gray Herbarium is also this species (Monte Verde, 

 eastern Cuba in 1859). 



Combs's 631, from Cienfuegos, in the Gray Herbarium is the type of C. eleusinoides 

 variety vestita. It differs from Grisebach 's type in being somewhat more pubescent. 



"Op. cit. pi. 49. 



