Mr Murray on the Course of the River Niger. 165 
tliird, or main stream, under the name of Quolla, rolls south- 
ward of east through Gauw, Zamfarra, Noofee, Boussa, and 
other countries, till, after §, long course, it also separates. One 
'branch rolls eastward, and, turning to the north, forms the 
Egyptian Nile ; the other flows southwards, and, again separa- 
ting, pours itself into the southern Atlantic ocean by several 
channels, of which the Congo is the principal. 
In suggesting some modifications upon this statement, it will 
be necessary to treat successively of the different lines of river 
course, delineated from this report of the African merchants. 
1. The Gambaroo. The existence of this river, and the fact 
that there are two great parallel streams, instead of one, running 
through the region east of Tombuctoo, forms a most important 
and unexpected accession to our knowledge of its geography 
There cannot, it should seem, be the smallest doubt as to there 
being such a river ; for all the merchants who give routes to 
Houssa, Kassina, and other countries north of the Niger, posi- 
tively state, that, after having crossed the main stream of the 
Niger, they come, in ten or twelve days, to this other great 
river. But, with regard to its being a branch separated from 
the Niger, and like it flowing eastward, there appears greater room 
for doubt. It may be first remarked, that this early and perma- 
nent separation of a great river into two branches, is a phenome- 
non very contrary to the general analogy of nature. Without 
inquiring into the circumstances in the structure of the globe 
which render it so, it need only be observed, that among all the 
multitude of known rivers, there occurs only one authenticated 
instance^ This is the canal of the Cassequaire, connecting the 
Orinoco with the Rio Negro, which is on too small a scale, and 
under circumstances too peculiar, to form almost any exception 
to the general rule. The improbability is much increased, when 
we find the same authority representing the Gambarra itself as 
immediately separating, and this excessively rare phenomenon 
as taking place twice within so short a space. 
* The fact discovered by Mr Bowdich, of De Lisle, in 1707, having delineated 
a river near Tombuctoo, under the name “ Gambarou ou Niger,” is very curious. 
The notice of such a river by D*Anville, which he refers to as mentioned in my 
woi^k, easts, as he supposes, only in delineation. I apprehend, however, that the 
G&mbra of Marmol is the Gambia,- which often bears that name in the old writers. 
