I)AMARAS, OVAMPO, AND NAMAQUAS 139
soon disappear, partly through evaporation, but principally from percolation through the sandy soil. Here and there a thin layer of less porous earth holds the water longer. The pool may then become sanded over, but water can be reached without trouble by digging and scraping. During a large part of the journey this looking out for signs of water and digging wells, after the first four hours' journey had been accomplished, was the almost daily occupation. The giving of drink to the oxen, three at a time, out of an improvised trench covered with canvas, into which the water was ladled, was a common feature at each encampment.
The digging for water was laborious. Sometimes the well was already dug by natives, but dry, and had to be so much deepened as to require a chain of three men to utilise it. One raised the water-vessel to another who stood a stage higher, and he to a third who stood breast high above the surface of the ground and poured its contents into the trough. On one of these occasions we had fallen fast asleep, dogs and all, utterly wearied, and found in the morning, to our astonishment, the tracks of elephants all about us. They had drunk at the well, disturbed nobody, and disappeared into the not distant bush, whither I followed them in vain.
The caravan at starting consisted of ten Europeans and about eighteen natives, or twenty-eight in all. The two wagons were both laden. The large one -had a solid deck over its cargo, and the space above deck was curtained into two compartments, in which Andersson and I slept when the ground was wet; as a rule we bivouacked in the open. The available