Going up through the village and palm gardens, and turning off in a northeasterly direction toward the desert, one presently comes out about midway of that valley to which I have made allusion more than once already. No one, however unskilled in physical geography, could look from end to end of that huge furrow and not see that it was once a river-bed. We know not for how many tens of thousands, or hundreds of thousands, of years the Nile may have held on its course within those original bounds. Neither can we tell when it deserted them. It is, however, quite certain that the river flowed that way within historic times; that is to say, in the days of Amenemhat III (_circa_ B.C. 2800). So much is held to be proven by certain inscriptions[54] which record the maximum height of the inundation at Semneh during various years of that king’s reign. The Nile then rose in Ethiopia to a level some twenty-seven feet in excess of the highest point to which it is ever known to attain at the present day. I am not aware what relation the height of this ancient bed bears to the levels recorded at Semneh, or to those now annually self-registered upon the furrowed banks of Philæ; but one sees at a glance, without aid of measurements or hydrographic science, that if the river were to come down again next summer in a mighty “bore,” the crest of which rose twenty-seven feet above the highest ground now fertilized by the annual overflow, it would at once refill its long-deserted bed and convert Assûan into an island. Granted, then, that the Nile flowed through the desert in the time of Amenemhat III, there must at some later period have come a day when it suddenly ran dry. This catastrophe is supposed to have taken place about the time of the expulsion of the Hyksos (_circa_ B.C. 1703), when a great disruption of the rocky barrier at Silsilis is thought to have taken place; so draining Nubia, which till now had played the part of a vast reservoir, and dispersing the pent-up floods over the plains of Southern Egypt. It would, however, be a mistake to conclude that the Nile was by this catastrophe turned aside in order to be precipitated in the direction of the cataract. One arm of the river must always have taken the present lower and deeper course; while the other must of necessity have run low--perhaps very nearly dry--as the inundation subsided every spring.