Frictional Resistance in Artificial Waterways 35 
Fig. 25. Siphon No. 17, Arkansas Valley Conduit, Pueblo. 
C 
of the depressions, but the high values o£ (n) obtained for some of 
the siphons indicate the. presence of debris or other unfavorable 
condition within those pipes. 
Experiments were conducted upon this series of siphons on 
August io and n, 1913. The structures had been in use since the 
spring of 1907. There are over twenty-five siphons on this canal, 
but with the method used in determining the loss of head, tests 
could not be made upon those having very turbulent condition 
at the inlet, due to the drop into the penstock or having high walls 
over which the water could not be siphoned with the hose. Fur¬ 
thermore, the flow into the conduit was practically constant which 
prevented the determination of the effect of velocity upon the fric¬ 
tion factors. Elevations of the water surface were referred to the 
company’s bench marks established upon the concrete inlet and out¬ 
let structures. 
Regarding the effect of ice in retarding the flow of the water 
in siphon No. 23, which is 50 inches in diameter, 2,797.6 feet in 
length, and under a maximum pressure head of 147 feet, Mr. R. 
M. Hosea, Chief Engineer of the Colorado Fuel & Iron Company, 
says: “The intake end of siphon No. 23 troubles us in winter time 
particularly by the formation of anchor ice inside the pipe, which 
appears to take place on the ascending leg of the pipe to such an 
extent as to check the discharge with a plug of ice particles. When 
the plug is forced through in advance of the water, on one or two 
