88 Floods on the River Banvon. 



about 150 feet.* How the water behaved at the gorge itself is 

 not accurately known, though one witness said it seemed to 

 be much lower in the middle of the river. It was at first 

 thought that the level (16*06) obtained from a well-defined 

 flood-mark made in 1880 in one of the rooms of the tannery 

 there, represented fairly the level of the water across the 

 gorge ; but further investigation throws much doubt on 

 this. The mark being at a sudden contraction, and at a 

 bend in the river, the level, as a mean for the section, should 

 (as before remarked) be accepted with caution. Owing to 

 the stream being confined at this place to the south side of 

 the river, there was probably as great a difference of level in 

 the flood on the two sides of the river as below the railway 

 embankment (see last par.). After much careful con- 

 sideration of the point, the author is of opinion that R.L. 

 15 00 is more likely to have been the average level of the 

 section than 16 '06. Calculations of the 1852 flood support 

 this view. 



14. A few words regarding the method of calculation 

 adopted may be useful to those wishing to follow up the 

 subject, or to apply the same method to other cases. The 

 ordinary rules and formulae are not strictly applicable to 

 such irregular channels as the portion of the Barwon River 

 dealt with in this paper. The formulae given in manuals 

 on hydraulics for irregular channels are, however, generally 

 so complicated that often, rather than use them, one prefers 

 to obtain the results by a series of approximations with the 

 ordinary formulae. In many cases, the results obtained by 

 either method must be looked upon as mere approximations ; 

 but it is better to accept them as such, than, because exact 

 results are not obtainable, remain in total ignorance of the 

 quantity of water to be dealt with, its velocity, and other 

 conditions of motion. 



15. In this case, the first thing done was to sketch on 

 the plan the probable course of the stream, and to take 

 the sectional areas of such portions only of the river as 

 would, under the special conditions of the case, be in 

 the stream. Next it was assumed generally that the 

 velocity varied inversely with the area in the stream. 

 Then, at each of the cross sections, the information required 

 has been calculated from the known data on the supposition 

 that the section is that of a uniform open channel. The last 



* Shown in the diagram by a dotted black line. 



