FLOW OF WATER IN PREDGED DRAINAGE DITCHES. 5 
the coefficient C is first obtained from Chezy’s formula, 
V=CVRS 
and then the value of n by solving for n in Kutter’s formula. In 
solving Chezy’s formula for C the following terms must be known: 
V, the mean velocity along the slope course. 
R, the mean hydraulic radius along the slope course. 
S, the mean slope of the water surface along the slope course. 
As has been explained, the data required for computing these 
values were obtained by field measurements. The mean velocity, 
VY, was determined by dividing the discharge by the mean cross- 
sectional area of the channel along the slope course. The discharge 
was computed from the soundings and velocity measurements made 
at the gauging section. The methed of computing the discharge 
was as follows: The mean velocity in the vertical at each velocity 
measuring point was obtained by taking one-sixth of the sum of the 
velocity at the surface, four times the velocity at mid depth, and the 
velocity at the bottom, or by taking the mean of the velocities 
obtained at 0.2 and 0.8 depths. The mean velocity for each section 
between the verticals was taken as the average of the mean velocities 
for the verticals on either side of the section, and the discharge for 
- each section was obtained by multiplying the mean velocity for the 
section by the area of the section. The total discharge of the stream 
was obtained by taking the sum of the discharges for all sections 
across the stream. 
The cross-sectional measurements were platted on cross-section 
paper, on a scale of 5 or 10 feet to the inch, depending upon the size 
of the channel. The areas and wetted perimeters for each cross section 
were obtained by means of a planimeter and map measurer. The 
mean cross-sectional area, which was used in determining the mean 
velocity for the course, was obtained by taking the average of all the 
cross-sectional areas along the course. 
For some of the channels the mean hydraulic radius for the course 
was obtained by taking the average of all the hydraulic radii as com- 
puted for each cross section; for others it was obtained by dividing 
the mean cross-sectional area by the average of all the wetted per- 
imeters as determined for each cross section along the course. There 
was no appreciable difference in the results obtained by the use of the 
two methods unless the channel was subject to large variations in 
cross section, in which case the first method was used. 
_ Onallof the channels the slope of the water surface was practically 
uniform from the upper to the lower ends of the slope courses, and 
the differences in the velocities at the upper and lower ends of. the 
courses were so small that the application of a correction for the differ- 
