been worked. At last we observed near the catacombs sundry heaps
of pinkish earth, evidently washed out; and our researches in the
South Country afterwards suggested that this may have been the
remains of the micaceous schist, whose containing quartz was so
extensively worked at Umm el-Harab. Moreover, a short study of
Shaghab threw more light on the matter.

Water also had been stored up with prodigious labour. We could
easily trace the lines of half a dozen aqueducts, mostly
channelled with rough cement, overlying a fine concrete; some of
them had grooved stones to divert the stream by means of lashers.
The Fiskiyyah or "tanks," as carefully built, were of all sizes;
and the wells, which appeared to be mediaeval, were lined with
stones cut in segments of circles: we shall see the same curve in
Sultan Selim's work near Ziba. The greatest feat is an aqueduct
which, sanded over in the upper part, subtends the left side of
the valley. It is carefully but rudely built, and where it
crosses a gully, the "horizontal arch" is formed of projecting
stone tiers, without a sign of key. This magnum opus must date
from the days when the southern part of the Wady was nearly what
it is now.

About a mile and a quarter below our camp, the Wady, which
broadens to a mile, shows on the left bank a wall measuring a
thousand metres long, apparently ending in a tank of 110 feet
each way. Around it are ruined parallelograms of every size,
which in ancient times may have been workshops connected with the
buildings in the island higher up. The torrents have now washed
away the continuation, if ever there was any; and, though the
lower remnants are comparatively safe upon their high ledge, the