In particular we know nothing about the origin of irrigation, which played so large a part in Egypt and Mesopotamia, and which has been repeatedly recognized as a factor greatly furthering social and political cohesion, since it makes each settlement dependent on its neighbours. We must, therefore, consider this invention. It deserves notice that irrigation can be resorted to by people who do not cultivate but collect wild-growing plants. This is done, for instance, by certain Indians of the Great Basin of Western North America,[54] and their methods could very well have been followed by the Natufians utilizing the wadi running at the foot of their cliffs. We may admit, then, that irrigation _could_ have been one of the features of the original agricultural complex which spread from the Near East; but there are serious arguments to the contrary. In the first place, the spread of agriculture seems to have been achieved by means of a slow migration of the cultivators. Primitive hoe or garden cultivation (which is still practised) exhausts the soil it uses. It ignores rotation of crops or fallowing; after some years a fresh piece of ground must be cleared and sown. When the neighbourhood has been farmed, the village moves farther into the bush. The smallness of the neolithic settlements and of their cemeteries,[55] and the manner in which they spread into the European continent, suggests this type of slow but continual migration outwards from the centre where agriculture was first practised. There is no dependence on irrigation to be observed here. In the second place, there are African parallels which suggest that the earliest agriculture in the Nile valley and Mesopotamia could also have proceeded without irrigation. The conditions in these river valleys in antiquity resembled closely those found nowadays on the Blue Nile, where semi-Hamitic nomads, the Hadendoa, sow and harvest in the simple manner which we shall now describe. It is possible, therefore, to postulate similar simple methods for the prehistoric Egyptians. Burckhardt renders his observations in the Taka country of Nubia as follows: About the latter end of June ... large torrents coming from the South and South-east pour over the country and in the space of a few weeks cover the whole surface with a sheet of water, varying in depth from two to three feet.... The waters, on subsiding, leave a thick slime, or mud, upon the surface, similar to that left by the Nile.... Immediately after the inundation is imbibed, the Beduins sow the seed upon the alluvial mud, without any previous preparation whatever. The inundation is usually accompanied by heavy downpours. The rains last several weeks longer than the inundation but they are not incessant, falling in heavy showers at short intervals.