314 Gleason: Studies on West Indian Vernonieae 



Vernonia divaricata. 

 Vernonia albicoma. 



2-10, 1906, in the Herbarium of the New York Botanical Garden. 

 Two other sheets in the same herbarium are Britton 151, from New 

 Haven Gap, and Shreve, from Sir John Peak. 



The seven species of the group may be distinguished by the 

 following key: 



A. Heads more or less secund, 11— 18-flowered; involucre cam- 



panulate to hemispheric, distinctly spreading when 

 press-dried; its scales loosely and irregularly imbri- 

 cated in few ranks. 



1. Principal leaves about 3 times as long as broad. 



a. Heads 11-13-flowered; pappus brown. 



b. Heads 18-flowered; pappus white. 



2. Principal leaves twice as long as broad, or less. 



a. Leaves broadly ovate-lanceolate, distinctly acu- 



minate; exposed portion of inner involucral 

 scales oblong; heads 18-flowered; cymes freely 

 branched. 



b. Leaves broadly ovate, acute or subacuminate; ex- 



posed portion of inner involucral scales ovate; 

 heads 11 -flowered; cymes sparingly branched. 



B. Heads not at all secund, 5-8-flowered; involucre narrowly 



campanulate or subcylindrical, even when press- 

 dried; its scales rather uniformly imbricated in 

 several ranks! 



1. Leaves spreading or ascending, nearly or quite sessile, 



distinctly oblong-obovate or subrhomboidal, 30-50 

 mm. long, acute; heads 8- (rarely 5-) flowered. 



2. Leaves proportionately narrower or larger, broadest 



near the middle, more or less acuminate. 



a. Leaves elliptical-oblong, about 25 X65 mm., con- 



spicuously reflexed; heads 8-flowered. 



b. Leaves narrowly elliptical-oblong, about 14X45 



mm., spreading; heads 5-flowered. 



Vernonia acuminata. 



Vernonia expansa. 



Vernonia pluvialis. 



Vernonia proclivis. 

 Vernonia reducta. 



Species-group Fruticosae 



When a sheet of Santo Domingan material came to hand, 

 identified by Dr. Urban as authentic V. fruticosa Sw., it was at 

 once seen that the species was entirely distinct from V. rigida 

 and from the whole group to which V. rigida belongs. It was also 

 seen that its nearest congeners were to be found among some un- 

 described species from eastern Cuba, with which it is accordingly 

 grouped. As in several other species-groups, the chief similarity 

 between the species included is in the general habit. The most 



