1102 
Journal of Agricultural Research 
Vol, V, No. 23 
stream side of the weir were those of the standard weir box—that is, 
the width of box was 10 feet; the depth of the box 6 feet; and the dis¬ 
tance of the floor from the crest of the notches about 4 % feet. A bulk¬ 
head was placed across the escape channel of the standard box, parallel 
to and about 5^ feet from the plane of the weir, thus making the spill 
box 10 feet wide, 5^ feet long, and 4 feet deep, the floor being about 2% 
feet below the crest of the notch. The height of the water in the escape 
channel was controlled by a steel head gate 20 inches square with a ver¬ 
tical slide set in the middle of the bulkhead about 0.5 foot above the 
floor, and by a 4-inch gate valve set near one end of the bulkhead, the 
finer regulation being made with this valve. The elevation of the water 
in the escape channel was determined by a hook gauge set in the concrete 
gauge box, which was connected with the escape channel by two i-inch 
pipes which entered near the floor line feet from the plane of the weir. 
Several minutes were required to adjust the flow of the water before 
an experiment was started, but when the desired condition of flow had 
been obtained it was maintained without difficulty throughout the test, 
except when the head on the upstream side of the weir was high and the 
head on the downstream side was small. Under this condition the large 
volume of water flowing through the notch depressed the water surface 
immediately downstream from the notch. This was followed by a 
standing wave, and the resulting backlashing and surging in the escape 
channel caused intermittent pulsations in the hook-gauge still box. The 
errors, however, were largely compensating, as is indicated by the con¬ 
sistent curves obtained from the experimental data. 
The discharges with different heads through the different notches, with 
free flow and with different depths of submergence, were plotted (figs. 
18 to 25) with discharges in cubic feet per second as abscissas and the 
heads upstream from the weir (Ha) as ordinates. Curves were drawn 
showing the discharges with different heads upstream from the weir 
(Ha) with varying differences (H D ) between the head upstream from the 
weir (Ha) and the head downstream from the weir (H B ). The method of 
interpolating between the values given on the curves in figures 18 to 25 
is indicated by the dotted lines in figure 18 and is based upon the fact 
that Ha = H b + H d . The H D = 0.15 line must pass through the points 
where the various H B lines intersect the Ha lines and satisfy the equation 
Ha — H b — 0.15. The H b = 0.65 line would be located similarly upon the 
points of intersection of the Ha and H b lines. Interpolations for other 
depths of submergence can be made in the same manner by drawing Ha 
lines for other than even 0.05-foot heads. For the purpose of compari¬ 
son, the free-flow discharge curve is drawn with each set of submergence 
curves. 
A series of experiments was made to determine the effect upon dis¬ 
charges of changing the conditions in the escape channel from free flow 
