HOLES IN RIVER BED 173 



would be interested in another person repeating it, unless 

 he did it for a wager. 



All this I now believe to be quite true, and the assertion 

 that I should not succeed in reaching the centre of the 

 continent I know to be so. Within a week of starting I 

 perceived that I should risk life if I persisted in pushing 

 on in a north-east direction with such a scanty supply of 

 food as that I had provided. I therefore turned due north, 

 determined to make for the coast and the settlements the 

 moment the food-supply threatened to fall short of our 

 requirement. 



We followed the course of the river-bed mentioned in 

 the last chapter for a distance of about sixty miles. All 

 the way we found the banks high, varying from forty to 

 seventy feet, while the bed itself was composed of white 

 sand with here and there streaks of a reddish-brown colour, 

 occasioned by some soil which had been washed into it. 

 The bed was full of holes, now containing water from the 

 recent storm-burst, and most of them were swarming with 

 fish of small size ; but as we had no means of catching any, 

 I cannot tell with certainty the species. Eels, however, 

 were amongst them, and a few of these were obtained from 

 the mud. Everywhere the sand of the bed concealed a 

 kind of light-coloured mud which held the water, and we 

 could obtain this in any quantity we wished by scraping a 

 shallow hole, into which it immediately flowed. All the 

 river-beds of this district are alike in this respect ; but as 

 they become drier, the water lies lower. It is only in 

 exceptionally dry seasons that it cannot be obtained at a 

 depth of seven or eight feet at most ; and I am convinced 

 that all the larger streams at least, have an underground 

 course, as is said to be the case with many of those in the 

 eastern deserts of America and other parts of the world. 

 That there is a large body of water under the beds of 

 these rivers is proved by the rapidity with which they fill 

 on the fall of rain, showing that they have not to absorb a 

 large quantity before becoming saturated. On the plains 

 the rain is absorbed much more rapidly ; indeed a fall of 

 less than two or three inches is not sufficient to form pools 



