THE FLOW OF WATER IIT CONCRETE PIPE. 33 
rough to the touch in the usual washed-cement pipe way. The 
discharge was measured over the rectangular, contracted weir shown 
in the plate (fig. 1). The crest is 3.02 feet long, the end contractions 
each 1 foot, and the bottom contraction more than 3 feet. For two 
observations the end contractions were very slightly less than twice 
the depth on the weir crest; for the others the end contractions 
exceeded twice the depth. Elevations of the water surface above 
the weir crest were determined with the hook gauge shown in the 
stilling box. The mean inside area of the pipe line was taken as the 
average area of 10 samples of the pipe remaining from construction. 
This area was 2.106 square feet, while the area of a true 20-inch pipe 
is 2.182 square feet. Thus this pipe is 3.5 per cent under size. The 
velocity of the water in the pipe was found by dividing the discharge 
in second-feet by the area of the pipe in square feet. This line was 
laid in 1911, thus being in its fifth year of operation at time of test. 
The coefficient C s = 0.313. 
No. 18, Experiment S-60. — 20-inch jointed reinforced concrete pipe, 
Temescal Water Co., Corona, Calif. — Water for irrigation and domestic 
use is conveyed through 27,000 feet of plain concrete pipe flow line 22 
inches in diameter and over 15 depressions which require inverted 
siphons. 1 Of these, 13 are of reinforced concrete, 20 inches in diameter. 
One of the longest of the siphons, about 4 miles from Corona, was 
chosen for experiment. Open standpipes occurred near both inlet 
and outlet of the siphon. Water columns attached to piezometer 
tubes of type A were used at both ends of the reaches chosen. For 
observations 1, 2, and 3, piezometer tube No. 1 was thrust 9.8 feet 
down the line from the standpipe at the inlet. For observations 4 
and 5 a hole was tapped in the top of the concrete pipe 101.7 feet 
from the intake manhole and the piezometer tube thrust 6 feet 
farther down the pipe. For all observations piezometer No. 2 was 
set in the pipe 0.8 foot upstream from a hole tapped in the concrete 
pipe 98.7 feet upstream from the manhole at the outlet. Velocities 
were ascertained by the use of solutions of potassium permanganate 
and of fluorescein. For observations 1, 2, and 3 the color was 
injected in the manhole at the inlet, while for observations 4 and 5 it 
was injected at the tap in the concrete pipe 101.7 feet from the man- 
hole. For all observations the color was observed at the manhole 
near the outlet. This pipe was cast on the ground in units 3 feet 
long. For moderate heads a 1 : 2 : 4 mixture of cement, sand, and 
gravel was used and for higher heads the mixture was changed to a 
ration of 1 : 2 : 3. All units were " washed " with cement grout before 
being laid. This siphon is straight in horizontal alignment but is on 
practically one long double-reverse vertical curve, the maximum head 
being about 80 feet, while the total length from manhole to manhole 
is 2,273 feet. The water is very clear and this siphon is undoubtedly 
absolutely free from debris, as all sand must pass more than 15 other 
siphons before reaching this one, the flow line between this siphon 
and the next one upstream being but 300 or 400 feet long. One 
blow-off valve is located at the low point of the siphon. The pipe 
waslaid early in the year 1911 and was thus about 5^ years old at the 
time of the experiment. So far as could be determined from the low 
standpipes, the joints were well smoothed and this pipe is probably 
1 Engin. Bee, vol. 64, Nov. 4, 1911, p. 526. 
164725°— 20— Bull. 852 3 
