68 ‘MR. SCOTT ELLIOT ON THE BOTANICAL RESULTS 
Wallace, this would be almost impossible to tabulate upon (Phil. 
Trans. vol. 183. p. 371), but his subdivision into Africa North 
and South of the Equator would separate both the Congo and 
Angola entirely from the Niger, which seems unfortunate. 
As to the origin of this Niger-Congo flora, we can at present 
only speculate. I have been able to see something of the 
Canary Island flora and of Algiers, and there does not seem 
any connexion with this Niger-Congo flora. The Sahara flora, 
known to me from Tripoli and Egypt, and the Upper Nile flora 
from Wady Halfa also show no affinity with it, neither does 
anything I have seen in Cape Colony, Transvaal, or Natal 
remind me of this district. On the other hand, Angola falls 
partly within it and there seems strong affinity with the Abys- 
sinia to Nyassa flora, shown, however, chiefly by genera not 
species. Hence it seems to me possible that the ancestors of 
these plants crossed into the Congo valley from the East, and 
were perhaps originally connected with India when the Nubian 
sandstones were deposited, z. e. before the arid character of the 
Sahara and Arabia was established. It does not seem at all 
likely that they came from the North, as the high altitude plants 
seem almost certainly to have done. 
In this connexion some have been inclined to suppose com- 
munication by islands from the West Indies to West Africa. I 
have been able to add a few more species common to both these 
regions (Symmeria paniculata, Hyptis, 2 species, Xyris laxiflorus, 
and Pycreus sp.), but we must remember the constant traffic 
which has taken place between West Africa and the West Indian 
islands almost since the discovery of the latter; both rice and 
maize, and possibly manioc, have been imported constantly for 
growing, and also tobacco. The ancestors of the present 
population of Freetown were also brought from America and 
carried with them, as is very obvious at Freetown, fruit-trees 
and probably medicinal plants, so that there has been great 
opportunity for accidental introductions; while the drift of 
the ocean currents seem to be mainly from Africa to America. 
On the other hand, the occurrence of such genera as Bertiera 
and Sabicea with many endemic species cannot be explained 
in this way, and a full knowledge of both Indian and American 
floras would be required before finally settling the question. 
