546 MR. G. BENTHAM ON COMPOSITA. 
American identical species are more specially in West Africa, 
accompanying a smaller number of endemic species and genera. 
There is, however, much yet to be discovered before the Com- 
posite of tropical Africa are sufficiently known to establish 
reliable proportions. Even the materials we have are not fully 
worked out, this part of Oliver’s Flora, although far advanced, not 
being yet completed ; and it is probable that the number of marked 
endemic forms, often connecting genera or higher races which are 
elsewhere distinct, will be considerably increased by future ex- 
plorers. We have, indeed, already indications of two species 
-which may have to be added to the endemic monotypes: one is 
the Moquinia Bojeri, DC., Prod. vii. 23, from Zanzibar, which, with 
its short corolla-lobes, is certainly out of place amongst Gochnatiex 
and may possibly be some Inuloid allied to Printzia ; the other 
is a fine plant with Eryngium-like prickly-toothed single leaves, and 
large almost Stifftia-like heads, of which we have seen several speci- 
mens in Afzelius’s herbarium, and one in the Kew herbarium 
collected by Barter on the Niger; in all these specimens the corolla- 
limb and upper parts of the anthers and style are eaten off or rotted 
away, so as to render the fixation of their affinities impossible ; 
but the anther-tails remaining in Barter’s specimen seem to in- 
dicate a new Mutisiaceous geuus allied to Dicoma. 
4. Tropical-Asiatic Region. 
This comprises the Indian peninsula, the plains and lower hills of 
India to the foot of the Himalayas, the island of Ceylon,the Malayan 
peninsula, and the Indian archipelago. Itis at once the poorest in 
Composite and the least diversified of all the continental regions, 
exemplifying on a large scale the characteristics of the low forest- 
lands of east tropical South America mentioned under the Brazilian 
region. Itis only the Nilgherry range, bordering the western side 
of the peninsula, that possesses a few endemic monotypes (Ade- 
noon, Lamprachenium, Nanothamnus, Athroisma) with restricted 
stations marking the site as a centre of preservation of detached 
races. The same hills have also endemic species of other genera 
of the same local character; but neither Ceylon nor any of the 
islands of the archipelago nor the Malayan peninsula, as far as 
known, have a single endemic genus of Composite, none more 
than a few species that do not also spread more or less over the 
Indian continent. A large proportion, indeed, of the true deni- 
zens (the Blumeas and allies, the Grangeinem, most Helianthoidee, 
