IIOUN EXPRDITION — NARRATIVK. 51 



dry it is .and may liavo \)oon pofliaps for montlis lioforo. Bot'oro tlio iii(ii'niii<i c-omcs 

 tlio formerly dry clianiiol may contain a roarin^^' torrent bearing ujirootcd trees and 

 shrubs along with it and spreading out far and wide over any low lying Hats, and 

 yoiii- baggage and impedimenta may be miles away down stream. 



The falling of the waters is almost as rapid as the rising ; as soon ,as the r.ain 

 ceases in the gathering ground the torrents from the rocks and hard soil dwindle 

 and dwindle until they disappear. The overflow level of the rock pools amongst 

 the ranges is .soon reached for there are no springs yielding a constant supply fi'om 

 waters which have been storefl up amongst the rocks. At most any small supply 

 of water oozing out of them serves only to balance the loss of evaporation so that 

 there is no surface flow. Very rapidly, when once the supply from the high lands 

 ceases, the water drains off', a large part fortunately siidiing down into the sand 

 where at some depth below the surface it slowly finds its way along. Where the 

 scour has been the strongest and has washed out the sand, there for a longer or 

 shorter period water will stand, prevented from sinkng into the saml by a thin 

 coating of the finer mu<l which, when the water is of some little depth and stands 

 quietly foi' a short time, will fall down and give rise to a thin rather clayey film on 

 the sui'face which will help to keep the water from sinking. Fortunately the 

 conditions are favourable to the formation of this impermeable film. When the 

 waters are at their highest the holes are scoured out in the sand ; as they fall and 

 begin to flow along sluggishly first of all the heavier sand particles are dropped, 

 lastly there remain only the finer mud p.articles with which the water, now foi-ming 

 only a thin surface layer, is heavily charged. As the surface layer flows over and 

 into the deeper pools, it brings into them a constant supply of fine mud until 

 gradually the flow ceases and the mud settles at the bottom and by its means a 

 supply of water is retained. 



Amongst other things, the flood brings with it flshes from the more permanent 

 sheltered pools amongst the ranges, and they survive as long as the water lasts. 

 Nowhere in the cei^tral region is there any evidence of the fishes having adapted 

 the habit of the Afi'ican mud-fish (Protopterus) to enable them to withstand a dry 

 period. There is indeed no Australian fish which can, so far as we yet know, exist 

 if the waters be dried up, and in the central region the reason for this is not far 

 to seek. The liver beds are sandy and after digging down ior some thirty feet no 

 material is met with out of which anything like a mud case could be made. 



Where the burrowing frogs are found by the side of water-holes and clay-pans 

 on the stony table-lands further south, no fish are met with. They could oidy 

 reach these water-holes in the form of eggs carried by birds, and long before they 



*4A 



