The Egyptian oases are deep and extensive depressions or hollows cut down nearly to sea-level through the generally horizontal rocks forming the Libyan Desert plateaux, and appear to owe their origin in great measure to the differential effects of subaërial denudation acting on rock-masses of varying hardness and composition. The surface-features or configuration of almost any land which has long been exposed to the powerful forces of erosion are more or less intimately dependent on the structure and lithological characters of the underlying rocks. On relative hardness, more than on anything else perhaps, depend the ultimate positions of mountains, hills, and plateaux on the one hand, of valleys, plains, and depressions on the other. Variation in the original conditions of deposition, at the time when the rocks now forming the Libyan Desert were laid down on the floor of the sea, has resulted in a preponderant development in some areas of soft clayey or sandy rocks (as compared with the hard limestones), and subsequent earth-movements have raised these beds more in some districts than in others. The result has been that wherever, during the gradual denudation to which the country since its elevation has been subjected, these soft deposits have become exposed on the surface, weathering has proceeded at a greatly increased rate, and eventually produced deep and broad depressions separated by high limestone tablelands. [Illustration: SKETCH MAP SHOWING THE GEOLOGY AND THE ANCIENT LAKES OF THE OASIS OF KHARGA] But for the presence of comparatively soft formations such as the Esna Shales, the Exogyra Beds, and the Nubian Sandstone, coupled with the facts that they have an unequal development in different areas, and occur at a greater elevation in some localities than in others, the great depressions of the Libyan Desert would not have come into existence, or at any rate would have been of comparatively little importance in the configuration of the country. The oases are true depressions, completely or partially surrounded by high escarpments. The oasis of Baharia, for instance, is on all sides hemmed in by cliffs; on the other hand, Dakhla and Kharga are open to the south, but as the ground in that direction rises considerably, they, too, cannot be regarded as other than true depressions. We have no definite grounds for considering that the erosion of these depressions can have been the work of previously existing rivers, and there is no evidence to warrant us in assuming them to have been formed by local subsidence of portions of the earth’s crust.