320 C. C. Brittlehank — Rate of Erosion of River Valleys. 



VII. — The Eate of Ekosion of some Eiver Valleys. 



By C. C. Brittlebaxk. Esq. 



(Read before the Geological Section of the Australasian Association for the 

 Advancement of Science, Melbourne Meeting, January, 1900.) 



THE district in which these observations have been made is 

 a very suitable one for such experiments, and is situated from 

 2 to 12 miles from Bacchus Marsh and from 32 to 54 miles west of 

 Melbourne. 



The chief streams vphich pass through the above area are the 

 Werribee and Lerderderg Kivers with their tributary creeks. The 

 country is somewhat rough, being intersected by numerous deep 

 gorges and wide flat-bottomed valleys. 



Early in 1892, when working on the glacial deposits of Bacchus 

 Marsh, I was greatly astonished at the immense amount of work 

 done by those puny streams, which have in places cut fairly 

 narrow gorges and valleys through solid rock, to a depth in on© 

 place of at least over 1,000 feet. 



It occurred to me that, if by any means I was able to ascertain the 

 amount of rock worn off the river bed within a given time, I should 

 be able to arrive at an approximate estimate of the time the rivers 

 have been cutting their channels. Even without detailed observa- 

 tion we should expect that a vast period had elapsed since the 

 streams first ran on the surface of the basaltic tableland. I was, 

 however, hardly prepared for the enormous time which my observa- 

 tions show, more especially as the erosion is sxibsequent to the flow 

 of newer basalt. 



After thinking the matter over, and striving to devise some 

 method other than the level, I hit on a somewhat rough and ready 

 plan, and it is from these measurements, taken over a period of five 

 years, that my provisional estimate has been obtained. My method 

 is as follows : — A number of holes are drilled in lines across the 

 river bottom — if in soft rock, one inch in diameter by three inches 

 deep, and one inch by two inches in hard dense rocks. Into these 

 holes are placed a number of wires which have been carefullj^ 

 measured, cut, and fastened by solder into small bundles containing 

 from 20 to 100 according to the hardness of the rock. The wires 

 being placed in the holes with the longest wire exactly flush with 

 the rock surface, cement is poured in and allowed to harden. As 

 various kinds of wire have been used in the softer rocks, a strip of 

 slate pencil of equal length to the longest wire is placed in each 

 hole ; this acts as a check on the wire in case of corrosive action. 



Now as the surface of the river bed is worn away the end of 

 a fresh wire is exposed for every hundredth part of an inch removed 

 from the harder rocks, and one for every fifth or twentieth of an inch 

 in the softer rocks. 



There is reason to suspect from geological evidence that the gorge 

 of the Lerderderg River above the alluvial flats is of much greater 

 age than that of the Werribee. Prior to the flow of newer basalt 

 the Lerderderg had cut a fairly deep channel, as had also the 



