for large vessels when Gordon reached Khartum in 1874.  It is practically impossible to keep the central provinces of the Nile open to civilisation unless the course of the Nile is free.  Yet in 1878 the obstruction had been renewed, and during the occupation of these provinces by the rebel dervishes under the Mahdi and the califa the Nile was completely blocked, as formerly, at Lake No.  The alarming failure of the Nile flood in 1899—­1900 was generally attributed to this blockade, and in 1899 fifty thousand dollars was placed at the disposal of the governor-general for reopening the White Nile by removing the vast accumulation of sudd which blocked the Bahr-el-Jebel from Lake No almost as far as Shambeh.  The work was started under the direction of Sir William Garstin in 1899.  In 1900 the greater part of the sudd had been removed by the strenuous labours of Major Peake, and the Nile again became navigable from Khartum to Rejaf.  The sudd was found to be piled up and of almost as close a structure as peat.  It was sawn out in blocks ten feet square and carried away by gunboats.  In the years 1901—­02 further progress was made, and twenty thousand dollars appropriated for the work; and by means of constant patrolling the sudd is now practically absent from the whole course of the White Nile.

The discharge of the flood waters from the Upper Nile begins to make itself felt in Lower Nubia and Egypt in the month of June, at first slightly, and after the middle of July much more rapidly, the river continuing to rise steadily till the first week in October, when it reaches high-water mark, nearly fifty-four or fifty-five feet at the Egyptian frontier, and twenty-five or twenty-six feet at Cairo.  A subsidence then sets in, and continues till low-water level is again reached, usually about the end of May.  The floods are then much higher and confined to a narrower space in the Nubian section of the Nile, while they gradually die out in the region of the Delta, where the excess seawards is discharged by the Rosetta and Damietta branches.  In place of the old Nilometers, the amount of the rise of the Nile is now reported by telegraph from meteorological stations.

It is popularly supposed that at every rise the plains of the Delta are inundated, but this is not the case.  The actual overflow of the banks of the river and canals is the exception, and when it happens is most disastrous.  The irrigation of fields and plantations is effected by slow infiltration through the retaining dykes, which are prevented from bursting by the process of slow absorption.  The first lands to be affected are not those which are nearest to the dyke, but those which are of the lowest level, because the waters, in percolating through under the ground, reach the surface of these parts first.  In Manitoba during a dry season sometimes the roots of the wheat strike down deep enough to reach the reservoir of moisture under ground.  In Egypt this underground moisture is what is counted upon, but it is fed by a special and prepared system, and is thus brought to the roots of the plants artificially.