328 THE DESERT SANDSTONE. 



At the Yam Creek section, I would remind readers that there is 

 ten feet of large water worn conglomerate, without any of the 

 characteristic sand. This has been clearly derived from the 

 granite before there was any admixture of other material. 

 Generally speaking the conglomerate increases towards the base 

 of the formation to the exclusion of the sandstone. 



But the curious fact observed in connection with these fluviatile 

 deposits is their great disproportion to the streams now connected 

 with them. At Yam Creek the valley is a mile and more in width 

 and the present small trickling stream is 90 feet below the bed of 

 conglomerate. Its waters increased tenfold would be utterly 

 inadequate to produce the erosion of 90 feet through granite and 

 a valley of such width. The same argument may be used towards 

 all the rivers in North Australia ; they seem so much smaller than 

 the deposits connected with them and the erosion effected that 

 one is forced to the conclusion that the rivers, though strangely 

 enough occupying the same valleys, have become reduced in an 

 extraordinary manner compared to what they were formerly. 



The erosion here referred to is different in its effects in the 

 different streams. Thus the valley of the Victoria River is 

 bordered on its whole course as far as I have seen by the fluviatile 

 conglomerate into which the waters of the stream are daily making 

 inroads. In the Daly River the same thing is taking place. In 

 the Katherine River which is the main branch of the upper Daly, 

 the river, as I have said, flows in some places through a wide and 

 rugged tract of sandstone conglomerate ; through other tracts in 

 a deep sandy valley ; and again, on the north side of a wide plain 

 and at the foot of a very gradual slope of Desert Sandstone. From 

 these considerations I think that the fluviatile sandstones are 

 never more than local deposits of limited extent. If the sandstone 

 once covered the whole of the country as Mr. Daintree supposes, 

 these rivers would cut through valleys of the formation. Instead 

 of this we have constant instances of the stream on one side 

 winding round a sandy slope, which is evidently the thinning out 

 of the deposit, while on the other side there is a wide plain of the 

 older formations without any evidence of their having formerly 

 been covered with an ash deposit. 



There is no very great difference between the conglomerates 

 connected with the fluviatile sandstone and what are called the 

 " drifts " of Victoria and other colonies. Mr. Selwyn divides 

 these into two formations, the newer drifts, and those which he 

 considers upper and middle gravels, boulders, and water-worn 

 conglomerates. The first are Pliocene or what Mr. Selwyn 

 considered to be Pliocene, and are auriferous ; the second he calls 

 Miocene, and have never been found to contain gold. In some 

 instances the oldest auriferous formation is made up of successive 





