558 MR. G. BENTHAM ON COMPOSITAE. 
Masafuera is known to possess at least one of the Juan-Fer- 
nandez Dendroserides. San Ambrosio and San Felix were visited 
in 1869 by Commander Simpson, of the Chilian vessel of war 
‘Chacabuco,’ who brought some specimens of the vegetation in a 
very bad state to Philippi, who has published them in the 
* Botanische Zeitung’ for 1870, p. 496. He mentions one Com- 
posita, which he proposes as a distinct genus, differing from the 
South-American Alomie by having pales on the receptacle, a 
character which can scarcely be admitted as sufficient considering 
its variability in some nearly allied genera. It is, however, judging 
from Philippi's description, evidently a distinct endemie species. 
A plant of which he has seen the foliage only appears to him to be 
very probably an additional species of the Juan-Fernandez 
Dendroseris. 
4. South-Sea Islands. 
The islands of the South Pacifie, from the Feejees to the Mar- 
quesas, have generally, in respect of Composite, the same character 
of extreme poverty observable in South-east Asia and the Archi- 
pelago. Scarcely twenty species are enumerated in the few works 
treating of their flora, the principal ones of which are Endlicher’s 
‘Enumeration,’ Guillemin’s ‘ Zephrytis Taitiensis, some of Asa 
Gray’s papers on the plants of the American Exploring Expedi- 
tion, and, still more recently, Seemann’s ‘ Flora Vitiensis.’ These 
twenty species include the weeds to be met with in almost all 
tropical and subtropical lists—Adenostemma viscosum, Ageratum 
conyzoides, Erigeron linifolium, Siegesbeckia orientalis, Eclipta alba, 
Bidens pilosa, Centipeda orbicularis, and Sonchus asper. There 
are also a Vernonia (Cyanopis), a Blumea, a Wedelia ( Wollastonia), 
and a Dichrocephala, belonging to the flora of the Indian archi- 
pelago and tropical Asia. Four species are described as endemic 
in the Fiji or in the Tonga islands—a Vernonia (Strobocalyz), a 
Blumea, figured by Seemann, both requiring, perhaps, some further 
comparison with Archipelago species, a Lagenophora, and a Glosso- 
gyne, perhaps both of them also in New Caledonia. There are 
also two species of Bidens described as endemic in the Society 
Islands, and allied to some of the Sandwich-Island species of 
Bidens or Coreopsis. Lastly, there is the very remarkable Fitchia, 
found once only, I believe, in each of two different South-Sea 
Islands, and systematically connected with none but the Juan- 
Fernandez Dendroseris, mentioned under the last head. Of the 
