454 PROCEEDINGS OF THE AMERICAN ACADEMY. 



Species rcducta vol e.vchisa. 



A. Armani (Balbis) Bak. in Mart. Fl. Bras. vi. pt. 2, 191 (1876). 



Eupatorium Armani Balbis, PI. Rar. Hort. Turin, 1810, p. 27^ 



t. 6 (1810). Orsinia Eupatoria DC. Prod. v. 104 (1830). Pi- 



queria Eupaiorium (DC.) Gardn. in Hook. Lond. Jour. Bot. vi. 



430 (1847). CUhadium rotundifolium DC. Prod. v. 505 (1836); 



Bak. 1. c. vi. pt. 3, 152 (1884), ubi syn. alia. = Clibadium Armani 



(Balbis) Sch. Bip. ex Bak. 1. c. (1884). 

 A. polyphylla (Sch. Bip.) Bak. = A. fastigiata (Gardn.) Benth. 

 A. spilanihoides D. Don ex Hook. & Arn. Comp. Bot. Mag. i. 238 



(1835) = Gymnocoronis spilanthoides (D. Don) DC. Prod. 



vii. 266 (1838). 



2. Revision of the Genus Ageratum. 



The genus Agcrntum L. has not l)een subjected to any general revi- 

 sion since its treatment in DeCandolle's Prodromus in 1836. For 

 many years it has been made to include plants of a considerable range 

 of habit and, what is more noteworthy, though chiefly defined by its 

 pappus, has been allowed to contain species of widely divergent 

 character in just this matter. While the more typical species have a 

 pappus of five distinct scales, others have a cup-like crown of very 

 short and connate scales; still others have been admitted into the 

 genus which instead of scales of definite number have short or long, 

 slender or slightly thickened, smooth or plumose bristles of indefinite 

 number ranging from 8 to 20 or more. Finally certain species have 

 been included from similarity of habit which possess no true pappus 

 whatever but merely a sort of annulus beneath rather than exterior 

 to the corolla. 



To render the genus properly natural and compact, as well as to 

 permit its more precise definition, it seems best to refer to Alomia the 

 species destitute of pappus, and to exclude also those species which 

 have a bristle pappus. The latter group consists of six South Ameri- 

 can species, namely A. Agrianthus Hoffm. (Agriaiithus cori/mbosus 

 DC), A. altcrnifolium (Gardn.) Bak., A. campulocUnioidcs Bak., A. 

 confertum (Gardn.) Benth., A. melissaefolium DC, and A. Pohlianum 

 Bak. At first, it seemed likely that these species could be appropri- 

 ately separated as a distinct genus. Schultz-Bipontinus seems to 

 have planned such a segregation in his undescribed Mclifisopsis. The 

 species, however, differ much among themselves, both as to habit. 



