Before concluding our remarks on the geology of the oasis we must not omit to call attention to the beautiful and varied fossil remains which are almost everywhere to be met with in the calcareous beds of the hills and escarpments. It is, of course, by the study and comparison of these organic remains that geologists are enabled to determine the relative ages of the beds in which they occur, and thus to correlate them with the rocks of other countries. The lower argillaceous and arenaceous deposits of the oasis are comparatively unfossiliferous. From any of the points of vantage, such as are afforded by the higher hills within the depression, the general level of the floor of the oasis does not appear to vary to any great extent, but actual levelling shows that this is not in reality the case; and it is this variation of absolute level which is the primary cause of the very varying volumes of water yielded by the artesian wells in the different districts. The average height of the centre of the depression in the neighbourhood of the village of Kharga is approximately the same as that of the Nile Valley plain in the latitude of Farshut. Ball, by comparison of a series of aneroid barometer readings with the barometric records for the same period at the Cairo observatory, deduced the value of a point near Kharga village as 86 metres above sea-level, and used this as his datum in calculating the levels of other parts of the oasis. Previous aneroid determinations of the same point had been made by Cailliaud (104 and 118 metres) and by Jordan (68 metres). But even when the greatest possible precautions are exercised, aneroid determinations, especially when made with a single instrument, are necessarily unreliable, and still more so when used for calculating the levels of different points on a plain having only comparatively slight irregularities of surface. Utilizing the figure obtained by the railway surveyors for a point near the termination of the line, we get values of 58 and 60 metres above sea-level for Kharga village and Bore No. 1 at headquarters respectively, and a bench-mark at the latter place, having a value of 60·1, is used as the datum from which all the heights given in this book are calculated. Unfortunately it is not possible, owing to the lack of sufficient check-levels, to state the limit of probable error, and it must therefore be understood that the value of our datum, which in the meantime may be accepted as the best obtainable, is subject to future revision.