The second possible other cause operating to diminish the static head of the artesian water of the oases is the progressive desiccation of a lake which may once have occupied a part of the Qattara depression. As mentioned on p. 110, the floor of this great depression, large areas of which are 80 metres and more below sea-level, is partly covered by a salt-marsh, which is so soft and watery that it can only be crossed at a few places. The hundreds of great water-cisterns cut in the limestones of the plateau to the north of the depression—cisterns most of which are now dry—as well as other ruins along the coast indicative of a considerable former population, seem to show that the rainfall in the littoral region has within the historical period been greater than it is at the present day. When the rainfall in the coastal region was greater, there must have been more drainage into the Qattara depression, and what is now salt-marsh was thus possibly once a lake of some depth. Assuming, as I think is likely, that an underground water-connection exists between the marsh occupying the bottom of the depression and the artesian water of the oases, it is obvious that any progressive lowering of the lake-level consequent on the change of climate must have lowered the static water-surface in the country extending southwards towards the oases. In the oases themselves the lowering of the static surface would of course be much less than at the lake; but it is quite conceivable that even in the oases the lowering may have amounted to the few metres which would cause some of the older and higher-lying wells to cease to flow. As both the deepening of the Nile channel in Lower Nubia and the desiccation of the Qattara depression are probably still slowly progressive, it is possible that these causes may to some extent account for any slow lowering of the static water-surface in the oases which may be still going on. 8. _“Lost” Oases—“Zerzura.”_