RELATING TO. AFRICA. 
37^ 
west, seems clearly to point it out as the true Nile, 
or Bahr-el-Abiad of the moderns. Proceeding west- 
ward from Egypt, Herodotus * knew with very to- 
lerable precision all the nations who inhabited the 
coast of the Mediterranean, as far as the Straits. 
He knew also the line through the Libyan de- 
sert, by Ammon (Siwah) and Aegila (Augila) 
to Fezzan, which Major Rennell has clearly recog- 
nized in the country of the Garamantes. From 
thence a chain of positions seems to carry us to 
that lofty point of the Atlas, which separates the 
plains of Morocco from Tafilet. Africa, to a cer- 
tain depth, was therefore pretty fully explored. 
His character of its three successive belts ; the first 
fertile and cultivated ; the second rude and inha- 
bited by wild beasts ; and the third an expanse of 
sandy desert, is perfectly correct and appropriate. 
The regions deeper in the interior were known 
to him only by the very short narrative of the ex- 
cursion of the Nasamones, which we have reported 
on a former occasion, t There seems considerable 
presumption, that the river flowing to the eastward, 
to which these travellers were carried, must have 
been the Niger. It has been suggested, indeed, 
that it might rather have been one of those rivers 
which descend from the Atlas, and water the plains 
of Tafilet or Sigilmessa. But if, as the localities 
f Lib. IV, 
f Vol. I. p. so-i. 
