STUDIES IN THE COMPOSITA. 181 
scarcely had a parallel in the history of systematic botany, 
he declined to assign a generic name to the group, for the 
reason that he seemed to see in—what has proven a com- 
‘plete enigma to all other botanists of the last seventy-five 
years—the Arnoglossum of Rafinesque,! a generic type that 
would naturally include these plants. 
Nine years after Cassini’s rejection of C. atriplicifolia from 
Cacalia, both De Candolle and Rafinesque, and perhaps in- 
dependently of one another, segregated the plants in ques- 
tion from Cacalia ; the former incompletely, indicating their 
characters, but elevating them to subgeneric rank only ;? 
the latter fully characterizing the genus MESADENIA, nam- 
ing the old species under this genus-name, and giving full 
diagnoses of several proposed new ones? But this New 
Flora monograph is not the original publishing of this 
genus as a genus. MtksabpENIA was sufficiently established 
by a paragraph in Loudon's Gardener's Magazine as early as 
1832* And, as for the Arnoglossum of the same author in 
1817, the name was a homonym, and could not stand, but, 
as I shall more fully indicate under one of the species, the 
type must in all probability have been one of the Mesadenia 
species. 
Still another specialist, in the study of the Composita— 
Schultze—in 1845 reaffirmed the proposition that these 
plants are not Cacalias, but placed them in Senecio;? a 
course which appears to have been first indicated by Sir 
William Hooker in 1833,° and which Mr. Bentham adhered 
to as late as 1873, though we must credit him with having 
had the consistency of remanding also to Senecio all, even 
the typical African succulent Cacalias; so that with him 
Cacalia as well as Mesadenia are but synonyms of Senecio.’ 
But this method of procedure, which herbarium botanists 
1 Fl. Ludov. 65. LEGEN vol. eer 
' 2 Prodr. vi. 329. 2 m. i. 
* New Flora, iv. 78, 79. ' Gen. PL iii. 449. 
* Vol. viii. p. 247. 
