Socially the Camerun natives stand at nearly the same low level of culture as the neighbouring full-blood Negroes of the Calabar and Niger delta. Indeed the transition in customs and institutions, as well as in physical appearance, is scarcely perceptible between the peoples dwelling north and south of the Rio del Rey, here the dividing line between the Negro and Bantu lands. The _Ba-Kish_ of the Meme river, almost last of the Bantus, differ little except in speech from the Negro _Efiks_ of Old Calabar, while witchcraft and other gross superstitions were till lately as rife amongst the Ba-Kwiri and Ba-Kundu tribes of the western Camerun as anywhere in Negroland. It is not long since one of the Ba-Kwiri, found guilty of having eaten a chicken at a missionary's table, was himself eaten by his fellow clansmen. The law of blood for blood was pitilessly enforced, and charges of witchcraft were so frequent that whole villages were depopulated, or abandoned by their terror-stricken inhabitants. The island of Ambas in the inlet of like name remained thus for a time absolutely deserted, "most of the inhabitants having poisoned each other off with their everlasting ordeals, and the few survivors ending by dreading the very air they breathed[286]." Having thus completed our survey of the Bantu populations from the central dividing line about the Congo-Chad water-parting round by the east, south, and west coastlands, and so back to the Sudanese zone, we may pause to ask, What routes were followed by the Bantus themselves during the long ages required to spread themselves over an area estimated at nearly six million square miles? I have established, apparently on solid grounds, a fixed point of initial dispersion in the extreme north-east, and allusion has frequently been made to migratory movements, some even now going on, generally from east to west, and, on the east side of the continent, from north to south, with here an important but still quite recent reflux from Zululand back nearly to Victoria Nyanza. If a parallel current be postulated as setting on the Atlantic side in prehistoric times from south to north, from Hereroland to the Cameruns, or possibly the other way, we shall have nearly all the factors needed to explain the general dispersion of the Bantu peoples over their vast domain. Support is given to this view by the curious distribution of the two chief Bantu names of the "Supreme Being," to which incidental reference has already been made. As first pointed out I think by Dr Bleek, _(M)unkulunkulu_ with its numerous variants prevails along the eastern seaboard, _Nzambi_ along the western, and both in many parts of the interior; while here and there the two meet, as if to indicate prehistoric interminglings of two great primeval migratory movements. From the subjoined table a clear idea may be had of the general distribution: MUNKULUNKULU NZAMBI