THE PLOW OF WATER IN WOOD-STAVE PIPE. 83 
recording gauges, which were frequently calibrated. While such records are 
not sufficiently accurate for the measurement of loss in short reaches of pipe, 
yet they are passably acceptable when the great length, 15,865 feet, is considered, 
especially for the higher velocities; at 10 feet per second the gross loss was 61.9 
feet from which was deducted 2.9 feet for entry and elbow losses. 
The plant was put in commission in November, 1908, and the tests made in 
1911. The pipe contains five steel elbows each of 15-foot radius where the curva- 
ture exceeds 20°. The total angle in these elbows is 317°. The loss at each elbow 
was measured by differential gauge and the total of such losses deducted from the 
gross loss before computing the net loss of head in the pipe proper. The original 
article does not state the method of measuring the quantity of water. 
No. 58. 90-inch Antelope Creek Siphon, Eastern Irrigation Block, Canadian 
Pacific Railway, Alberta, Canada. Continuous Stave Creosoted Douglas Fir 
Pipe. 1 — In 1921 G. F. P. Boese and C. M. O'Neil made a test of the losses through 
a new stave siphon 1,698 feet long that replaced three barrels of a five-barreled 
monolithic concrete siphon. The new pipe, of pressure-creosoted fir staves, was 
designed by our formula, extended beyond any experimental velocities. For a 
required capacity of 700 second-feet the velocity in a 90-inch pipe would be 15.84 
feet per second. At the time of test the flow in the canal" was 639 second-feet, 
so a very satisfactory high- velocity test resulted. The mean velocity of 14.48 
feet per second was determined by traversing the pipe across horizontal and 
vertical radii with a pitometer. The instrument was held at definite points that 
determined the velocities at definite ring areas. Unfortunately the pipe diameter 
was so large that the pitometer rods could not reach to the center of the line by 
183^ inches in the vertical and 2034 inches in the horizontal. Thus it was neces- 
sary to project the velocity curve to the center. This is in a measure verified 
bj' the pipe coefficient found; that is, the mean velocity divided by the center 
velocity was found to be 0.875, which agrees fairly well with accepted coeffi- 
cients for pitometer work. 
The high velocity, 14.48 feet per second, makes this experiment of particular 
value as indicating that our formula may be extended beyond the range of the 
base data. The tables in this bulletin were computed for such high velocities, 
but at this time it was appreciated that the base data were being extended be- 
yond experience. 
The results of this test agree with our formula within about 4 per cent. 
No. 60. 96-inch Continuous-Stave Douglas Fir Penstock. Searsburg Pipe 
Line, New England Co. Power System, Vermont. 2 — In 1923 the engineering 
force of the Power Construction Co., under the direction of A. C. Eaton, made a 
series of tests on plant efficiency under the supervision of C. M. Allen, of Worcester 
Polytechnic Institute. These tests comprised a series of experiments on an 8-foot 
stave pipe line laid in 1921. The line consists of 18,406 feet of stave pipe plus 
seven scattered sections of steel special bends, aggregating 138 feet; also one 
concrete section 98 inches in diameter, 126 feet long, with additional concrete 
transition sections totaling 22 feet in length. The general alignment is quite 
sinuous. The velocity in the pipe was determined by the salt velocity method, 
which is used in much the same way as color except that the first and last arrival 
of the salt injection is determined by the deflection of an ammeter rather than 
by visibility. This method was used by Mr. Yarnell of the United States Bureau 
of Public Roads at Arlington, Va., in 1916 and was made public by Professor 
Allen in 1922. The loss of head was determined from readings of a staff gauge in 
i The results of this test were submitted in correspondence to the -writer by A. S. Dawson, chief engineer, 
Irrigation Block, Canadian Pacific Ry., Calgary, Alberta. 
J The results of this extended series of tests were submitted in correspondence to the writer by A. C. Eaton, 
hydraulic engineer, New England Power Construction Co., Worcester, Mass. 
