380 
THE NEW AFRICA 
branch track joins the main road again. We outspanned here, 
intending to travel for all we were worth in the cool of evening 
and the night. But when we resumed the journey we found 
a little pan of beautifully clear water four miles ahead, which 
delighted us with its unexpected appearance and freshness. A 
heavy thunderstorm had evidently passed here a few days 
previously, and filled this little hollow with water. In a very 
short time we had emptied out all our Lichachane fluid and 
replenished our small casks, bottles, etc., with the beautiful 
rain-water, and, to make matters easy, decided to rest another 
day here, after giving our oxen a good drink that evening. 
Through some strange presentiment that all was not right, I 
woke about midnight and walked over to the pan, and dis¬ 
covered that nearly all the water had disappeared. I gave the 
alarm at once, and before long we were all digging a trench 
in the soft sand with our hands to collect the little remaining 
water; we then drove the oxen up for a drink. Without 
further delay we inspanned and trekked on. This opportune 
pan had helped us eight miles over our difficulty. 
These pans are the result of a collection of rain-water in 
some chance hollow or depression in the desert, lined at the 
bottom with a coating of mud that has been washed out of the 
neighbouring soil, and deposited in sufficient quantity to form 
an impermeable layer strong enough to hold water if left undis¬ 
turbed. However, when perforated by such agencies as the 
feet of oxen, or by means of a foolish attempt to dig a well to 
concentrate the water, the continuity of the mud layer is dis¬ 
turbed, and the water leaks out into the underlying sand, 
thereby completely frustrating any attemps to ensure a better 
supply for a length of time. The effect of breaking the mud 
layer is the same as if so much water had been spilt in the 
sandy desert, and one vainly tries to collect it again. 
Some of the larger pans, such as Haakedoorn Vley and 
Nkawana, possess a thicker layer of mud at their bottoms, 
which in a moderate measure withstand the perforating action 
of oxen’s feet, and consequently are considered as more or less 
