558 PROCEEDINGS OF SECTION H. 



results of my enquiry to the Association, in the hope that the 

 facts may be of use, whether the references be so or not. 



Rankine warns the engineer not to embank across flooded 

 land unless assured that it acts as a mere reservoir, and as a 

 channel for the flood waters, but does not tell how the engineer 

 is to ascertain this fact if he does not actually happen to have 

 witnessed the flood.* 



Mr. Stanley, Chief Engineer bf Railways of the Southern Dis- 

 trict of Queensland, recommends that an allowance be made for 

 one inch of rain per hour, with an addition of twenty-five to 

 thirty per cent, for contingencies. He does not state what limiting 

 velocity he proposes through the opening, but as he allows an 

 eighteen inch pipe for eight acres that may be computed. Accord- 

 ing to him then, the waterway should vary directly as the area 

 drained, and should give from one hundred and twenty to one 

 hundred and forty square feet per square mile of country. The 

 velocity works out as seven feet per second, which even in very 

 flat districts could be obtained without heading up the water more 

 than a foot or fifteen inches on the up-stream side.f 



Mr. M. E. Kernot, of the Victorian Railways, who has given 

 this subject much attention, writes that a waterway of sixty square 

 feet per square mile in steep and rocky districts, forty square feet 

 per square mile in average undulating country, and twenty to 

 thirty in very flat sandy areas works well in those parts of Victoria 

 where the rainfall does not exceed thirty inches per annum for 

 catchments not exceeding three square miles ; and other Victorian 

 engineers have verbally informed me that their practice roughly 

 agrees with this. 



A formula given by Mr. Cleeman as having worked well on the 

 Richmond, Fredericksburg and Potamac Railway, U.S. America, 

 reads as vollows : — 



Area of waterway in square feet = C v' area drained in acres, 

 area drained in acres, where G varies from 1 to 1.6 according to 

 ciicumstances.| In Jackson's "Hydraulic Manual," (p. 20), a 

 similar formula occurs, in which the discharge from a given 

 catchment varies as the square root of the area. 



Mr. G. R. B. Steane, C.E.,§ gives the following : — 



Area of waterway in square feet = (area drained in acres,) ■*'^ 



as suitable for tolerably impervious surfaces, but stigmatizes it as 

 " only a rough approximation." Others have thought that the 

 discharge and catchment should vary as the two-third power of the 

 area drained. 



* Civil Enaiineering', p. 718. 



t See Roval Commission on Railway and Public Works of Tasmania, 1886, pp. 9, 65, 67. 



I See Proc. Inst Civil Engineers, Vol. LVIII., p. 375. 



§ Trans. R. Soc. Victoria. 1887, p. 154. 



