Between the camp and the well, which we easily found, there ran a well-beaten foot-pad, showing that this had been a favoured spot for some time past.  The well itself was situated in a belt of mulga-scrub, and surrounded by a little patch of grass; growing near by, a few good camel bushes, such as acacia and fern-tree (quondongs, by the way, were not seen by us north of Alexander Spring, with the exception of one near McPherson’s Pillar); enclosing the scrub two parallel banks of sand and stones, with the well in the valley between.  Above the well, to the, North, high anthills and tussocks of coarse grass appeared.  The whole oasis covered no more than three acres.  The well itself resembled those already described, and appeared to have a good supply, so much so that we started at once to water the camels, which had had no drink since August 21st, a period of seventeen days, with the exception of two gallons apiece at Warri Well, where the parakeelia grew.

By midnight all but three—­Satan, Redleap, and Misery—­had drunk as much as they could hold.  These three had to be content with a small amount, for we could not get more without digging out the well, and this we proceeded to do.  The night was hot and cloudy, and constant puffs of wind made work by the light of candles so impossible that we had perforce to bear the extra heat of a blazing fire.  The native well, as we found it, had been scooped out with hand and cooliman, just large enough to allow one to descend to a depth of fifteen feet, and the sides of the hole plastered back with mud, which had baked hard.  To follow this hole further was not feasible, for going down on a slope as it did, any further deepening would cause the sand to fall in; we had therefore to start a new vertical shaft from the surface.  After a considerable amount of digging we reached water level, and were preparing to bail the water, when with a thud the whole thing caved in, and our labour had to be recommenced.  At the time the wedge of ground fell in Godfrey was working below and narrowly escaped being buried.  A timely rope fortunately saved him.  I never saw a man come quicker out of a hole!  Now we were a bit puzzled.  Our position was this:  six camels were watered, three were not, our tanks were empty (my fault, for I should have first filled them and then the camels; but yet if we had water and the camels had none, would we have been better off?); our well, containing X, an unknown quantity of water, had fallen in.  Query, whether to recommence digging, or to pack up and follow the blacks?  Now, the well might contain a good supply, or yield no more than a gallon or two; and the blacks might or might not have gone on to a good water.  It was a puzzle.  Finally we compromised, and I sent Breaden and Warri to hunt up the tracks, whilst we started work again.  On one side of the well was rock, and by strengthening the other by timber we hoped for success.  Luckily plenty of good mulga trees were handy, and we soon had the timber ready for use.  This was the second night without rest or food, and no more than a mouthful of water each, for on arrival we had given what our tanks contained to the thirsty camels.