78 BULLETIN 852, U. S. DEPAKTMENT OF AGRICULTURE. 
The quantity was taken as the mean indicated by two current 
meter tests, one giving 3.19 and one giving 3.65 second-feet. This 
agreement is not as close as it should be. For this reason the writer 
would not give too much weight to the results of the experiments on 
this pipe in considering the friction loss which appears rather low 
for this type of pipe. The pressure head was measured by a " testing 
gage." The one observation on this pipe shows a value of C s = 0.314. 
No. 19, Experiment. — Fanning 20-inch cement-lined wrought-iron 
force main pipe. — In 1880, J. T. Fanning conducted a series of experi- 
ments and listed the results in his well-known work on hydraulics, 1 
but he does not give a description of the pipe or of his methods of 
experimentation. Hering and Trautwine say of this pipe: 2 
No short bends, but two large Y branches, two small blow-off branches, and three 
stop valves. 
No. 24, Experiment N-3. — 30-inch reinforced concrete jointed 
pipe, T> x siphon, Umatilla project, United States Reclamation Service, 
Oregon. — This pipe, laid in the winter of 1909-10, was tested for 
capacity by Mr. Newell in 1911 and by the writer in 1915. (See 
No. 23.) For general data on the Newell tests see No. 12a. 
The elevations of the surface of the water at the inlet and outlet 
structures were determined by the use of a hose pipe and water pail, in 
a manner similar to that later used by the writer when other methods 
were not feasible. By means of a true siphon over the wing walls the 
surface of the water in the pail and in the canal are brought to the 
same level. 
For observation No. 1 a single weir measurement was accepted as 
indicating the discharge, while for observation No. 2 a single meter 
measurement indicated a discharge of 16.62 second-feet and a meas- 
urement at the weir at the inlet indicated a discharge of 17.04 second- 
feet. As noted in Table 2 the various elements have been computed 
for both these indicated discharges, rather than averaging them. 
This method follows Mr. Newell' s original article. 
A comparison of the platted points for this pipe (PI. VI) indicated 
that it was very smooth when new, but had become somewhat ob- 
structed when tested by the writer four years after Mr. Newell' s tests. 
The relationship of the leading canal to the intake clearly shows that 
this is highly probable. The average value of C s for the three observa- 
tions equals 0.408, which is higher than usual, even for the best of 
construction. 
No. 25, Experiment N-4. — 30-inch reinforced concrete jointed 
pipe, E 3 siphon, Umatilla project, United States Reclamation Service, 
Oregon. — This pipe was tested by Mr. Newell when it was in its 
fourth season of service. The discharge was taken as the mean of 
three measurements by current meter, ranging from 12.76 to 12.42 
second-feet, to which was added 0.032 second-foot which passed a 
weir. Mr Newell states: "It is not unlikely that there is a con- 
siderable deposit of sand in the first low depression/' This would 
not appear to the writer to be a serious deposit, as the friction loss 
indicates a very efficient pipe. The relatively great length of the 
reach, 3,658 feet, makes this a valuable test, so far as one observation 
is indicative. The value of C s equals 0.351. 
i A Practical Treatise on Hydraulic and Water Supply Engineering, by J. T. Fanning, 11th ed., New 
York, 1893, p. 238. 
2 E. Ganguillet and W. R. Kutter, translated by Rudolph Hering and John C. Trautwine, jr. A General 
Formula for the Uniform Flow of Water in Rivers and Other Channels. New York, 1907, 2d ed., p. 154-155. 
