THE FLOW OF WATEE IN IRRIGATIOlSr CHANNELS. 61 



may entirely change the character of the bottom of a smoothly lined 

 canal. On the other hand, a water laden with fine silt flows far more 

 freely after the silt has deposited in a sHck coat over minor irregu- 

 larities than in a new, though clean, canal. A canal carrying such a 

 water may be designed for a far higher velocity through the same kind 

 of soil than would be the case if the water were clear. It is necessary 

 only to run but low heads in the new canal until a thick waxy deposit 

 has been placed on the canal bed, after which the velocity may be 

 nearly doubled over that which would have scoured the material in 

 which the canal was originally excavated. (/) The prevailing wind 

 direction may be given some consideration. A study of vertical 

 velocity curves shows a marked change in form with change in wind 

 condition. A downstream wind aids the flow of surface water to 

 the extent that it has the maximum velocity in the vertical, while an 

 upstream wind so shapes the velocity curve that the surface velocity 

 is as slow as that near the bottom. 



(4) That there is a tendency toward a lower value of n as the ve- 

 locity and hydrauhc radius increase. Any experiments that are in- 

 tended to bring out conclusively the extent of the variation and 

 whether it is due to the change in velocity or the change in hydraulic 

 radius must be conducted m very long, straight channels, in wood, 

 or concrete, or steel, where the character of the wetted perimeter will 

 remain unchanged as the water becomes deeper and deeper in the 

 channel. Proper corrections must be applied for any slight changes 

 in the mean velocity at the two ends of the reach tested. The reach 

 must be far removed from all influence of curves and structures. 



(o) That a value of n must be chosen that will apply to the canal 

 in question at the critical period of the season. For instance, most 

 canals are cleaned once a year. A growth of moss may become very 

 heavy by July or August, but the water supply or demand wiU prob- 

 ably be much less than during the early days of June. If the canal 

 is designed to carry its peak load on the basis of its being in good 

 condition, there; wiU still be sufficient carrying capacity for the smaller 

 discharge when moss has appeared. 



(6) That in the design of earth channels having a trapezoidal form 

 when constructed, the value of K should be computed on the basis 

 that th(! canal takes an elliptical form within a short time and there- 

 after maintains this shape unless altered artificially. 



acknowledgmp:nts. 



Tlie uTitcr and his associates desire to acknowledge indebtedness to 

 the various ciigiiu'ers and rniuuigcrs of irrigation and power systems 

 who permitted tliC use of canals under their cJuirgc; for the tests and 

 to the engineers of the United States Reclamation Service who al- 

 lowed access to much original data on tests mmlr, by their engineers. 



