ON THE MIGRATIONS OF RACES. 323 



the negro races in the so-called Soudan, the Fellahs inclosed between 

 the negroes and reaching from east to west in a straight line, and the 

 central races spread from the north and northeast to the equator. 



Of these five races, the first four only can be regarded as autoch- 

 thonous, while the last, comprising well-established groups, migrated 

 from Asia. 



The Hottentots were formerly the exclusive inhabitants of south- 

 easterly parts of Africa, from the Cape up to 18° or 19° south lati- 

 tude. They were driven from their settlements by the invading 

 Caffres streaming from the north, and at first were pressed back into 

 the most southerly regions, and then later from this extremity north- 

 ward along the west coast until they fixed themselves in the districts 

 they now occupy. The northern neighbors of the Hottentots, the 

 Caffres, are not aboriginal in the southern country, where at pre- 

 sent they most numerously exist, but have immigrated here. They 

 settled originally farther north, and stood in close relations for a long 

 period with the Hamitic peoples, which migrated from Asia, as is 

 clearly shown by their idioms. Since by reason of their type and 

 intimate relationship they could not, for any length of time, have been 

 separated by their primitive languages, which were continually approxi- 

 mating, they may have formed an individual group at the time of the 

 invasion of the Hamites from the north into Africa ; but they exhibit, 

 in fact, so close points of resemblance with the Hamitic idioms that, 

 without attributing this to direct contact, the coincidence appears inex- 

 plicable. Besides this drift from the north to the south, which is 

 established from already ascertained facts, another from east to west 

 diagonally across the continent was later instituted. From this cir- 

 cumstance it happens that the language of many stocks in the extreme 

 northwest of the Caffre area show the most intimate relationship with 

 those of the extreme northeast — a relationship not to be accounted for 

 by a reference of both to the primitive tongue common to the Caffre 

 or Bantu tribes, but completely through derivation from a branch of 

 this original speech. 



That the Fellah races are not aboriginal in those regions which 

 they at present occupy is proved by their distribution among the 

 negro races. Such a stratification of two races can not be aboriginal, 

 but indicates distinct migrations of each. According to our view, the 

 Fellah originally settled north of the negroes, probably in the territory 

 now possessed by the Berbers, and pressed from the northwest into 

 the land occupied by them, whence they spread toward the east to 

 Nubia. This opinion is confirmed by the close relationship of the 

 Fellah races with the central tribes, which appears to demonstrate an 

 intermixture, as also by the many points of resemblance which the 

 Fellah idioms offer to the Hamitic tongue. 



That the individual peoples into which the negro race is divided 

 have undertaken many migrations is at the very outset established by 



