14 BULLETIN 376, U. S. DEPARTMENT OF AGRICULTURE. 
2. That the mean of all reliable observations on carrying capacity 
of wood-stave pipes agrees with the exponential formulas in the fol- 
lowing order and per cent (see Table 9) : 
Per cent. 
1. Scobey +0.19 
2. Williams-Hazen 1 +4. 7 
3. Tutton +3.6 
4. Moritz -9.4 
3. That the mean of the capacities of the several pipes agrees with 
the exponential formulas in the following order and per cent (see 
Table 3) : 
Per cent. 
1. Scobey +1.29 
2. Williams-Hazen +4.96 
3. Tutton +7.25 
4. Moritz -7.72 
4. That Kutter's formula with a constant value of n does not 
apply to flow in wood-stave pipes running full. 
5. That n decreases with an increase in velocity in a given size of 
pipe and increases with the size of pipe for a given velocity, varying 
from less than 0.010 for small pipes at high velocities to more than 
0.014 in large pipes. 
6. That this variation in n is so marked and complicated as to 
render the use of Kutter's formula inadvisable. 
7. That the Ogden experiments showed the capacity of the 72-inch 
pipe (Nos. 47 and 48) to be within from 5 to 8 per cent of the average. 
8. That the Sunnyside experiments showed the 55f-inch pipe (Nos. 
45 and 46) to be abnormally smooth by 18 per cent. 
NECESSARY FIELD DATA FOR DETERMINING THE RETARDATION 
ELEMENTS OF VARIOUS FORMULAS. 
A glance at pages 5 to 7 shows that for study of the various formu- 
las the same hydraulic elements must be determined by field tests. 
These are : 
1. The mean velocity, V, of water in the pipe. 
2. The loss of head, h f , due to retardation in a section of pipe of 
uniform size, within a known distance. 
3. The internal size of pipe, D or d. 
The above data having been secured, the observed velocity for any 
particular observation may be compared with the computed velocity 
for the same-sized pipe with the observed loss of head, for any of the 
formulas. 
MEAN VELOCITY OF WATER. 
The velocity of the water flowing in a section of wood-stave pipe 
may be measured in two general ways : 
i Using coefficient of 120 in Williams-Hazen formula. 
