1. 2. XXXV. AND THE LIBYAN ELEMENT 29 opposite direction at other times , and I am strongly inclined to think that the NoBATAE were once lords of the Bayuda and the country south of it and that their negro ancestors may have previously ousted the Libyan races therefrom, or, more probably, become fused with them in race.

XXXIV The nehes of Senusert III and Thothmes I, and, to a modified extent, most of the western and southern subjects of the Nubian dynasty which ruled Egypt, may have been of the same stock as these negro ancestors of the Nobatae and their partly Libyanized descendants respectively; and if the racial substratum was the same, it is natural to suppose that the extent and direction of migration to and from the river were at different periods regulated by the weakness or strength, as the case might be, of other and hostile races living on the Nile.

For instance, when the tide of negro invasion was rolled back from northern Nubia in the time of the New Empire, most would retire southwards along the river, but the presence of alien Nilotic negroes in their path, or other causes, may have led a proportion to move westwards, where their kin may or may not have been already established. If the same stock, as modified by Libyan admixture, was strong enough at a later date to support the Nubian Empire of Napata it was by then probably predominant also in northern Kordofan, and a steady intercourse would naturally exist between the riverain and western groups, until the third century B.C. when Eratosthenes spoke of them collectively as NovfSai.

In any case there is evidence of a close connection between the negro invaders of the second millennium B.C., the rank and file of the Nubians who conquered Egypt under the twenty-fifth dynasty, the inhabitants of the Gezira at the same period, the present inhabitants of the Nuba mountains of southern Kordofan, the Nobatae of Lower Nubia, and the so-called Nuba of northern Kordofan.

XXXV The weakest link in the chain is perhaps that connecting northern and southern Kordofan, but even here, at the present day, though the general physique is obviously quite different, there is a common fund of superstitions connected with rainmakers and serpents, and recognizable cranial resemblances.

Particularly the flattening of the fronto-parietal region. This was pointed out to me in 1912 at Kaga by Professor Seligman (q.v. in Harvard African Studies, 11; Varia Africaiia, il, 181). It may be worth mentioning in this connection that some years ago I was told at Kaga that the ancient ("Anag") population used to bury their dead upright.' Lepsius was told the same of Southern Kordofan {Discoveries..., pp. 221-2). Kordofan takes its name from the hill of that name close to el Obeid. As used by the natives the name does not properly apply to the nomad country to the north nor to the Kaga-Haraza group of hills adjoining it nor to the Nuba mountains in the south. See MacMichael, Tribes..., App. I.