The total annual discharge of the whole of the wells of Kharga Oasis is barely equal in volume to the water which can be held by saturated beds underlying 1 square kilometre of surface, assuming the sandstone to be only 122 metres thick; that is to say, it would take between 3,000 and 4,000 years for the existing wells to discharge the water held by the beds underlying the depression alone, without considering the vast surrounding desert areas, where there is no reason to doubt that the water-tables are equally well developed. It must not, however, for a moment be supposed that bores would continue to discharge until the sandstones immediately surrounding them were completely depleted of water; they would, in all probability, pass into a sub-artesian condition in a very short time were the sandstones not replenished from more outlying districts. In my opinion the subterranean water of the oasis is certainly of meteoric origin—that is to say, it is water which originally fell as rain, and has percolated underground from one of the possible sources above mentioned. It will, however, readily be admitted that the ordinary explanation of the flow and origin of artesian wells in regions of moderate or abundant rainfall, situated in well-defined basins, where the exact position, extent, and absorbing capacity of the water-table outcrop can be carefully determined, may be in some respects inadequate to account for the flowing wells of vast arid regions like the deserts of Africa and Australia; yet, after due consideration of the arguments for and against, I am unable to subscribe to Professor J. W. Gregory’s view[11] that a considerable portion of the waters in such regions is derived from magmatic or plutonic sources—that is to say, has its origin in the deep-seated crystalline rocks.