Divisors 
5 
is the difference between the “head” in the main ditch and the 
depth of water in the “divisor” taken 3 feet downstream from the 
point of the divisor board. 
In the experimental work the divisor board was given a bevel 
of 1 to 4 on the “divisor" side, leaving the end about ]/% inch 
thick. Rubber packing was placed under the divisor board and the 
board screwed to the floor of the box to insure against leakage. 
The flow through the “divisor” and “channel” was measured volu- 
metrically in every experiment. 
As used in the field, sometimes a drop is placed in the ditch 
immediately below the box, depending on which, if either, of the 
ditches has an excessive grade. The ditch with the drop will have 
the “fastest” flow, as it is commonly expressed, and, therefore, 
get the lion’s share of the water. Since the conditions in the two 
ditches may be such as to produce a combination of velocities 
varying from slow to fast, it is evident that the divisor box can¬ 
not be classed as an accurate measuring device, and this is proved 
in the accompanying tables. Different types of boxes, settings 
of divisor board, depths of water and conditions of flow in the 
two ditches, are given in the tables with the hope that irrigators 
may know some of the facts connected with their own divisor 
boxes and apply the information. 
A comparison of columns 2 and 3 in the tables will give the 
error caused by assuming the discharge to be proportional to the 
distances from the point of the divisor board to the sides of the 
flume; column 4 gives the per cent of the flow in the main ditch 
which will flow through the “divisor” for the corresponding width 
of “divisor opening” shown in column 5; and columns 6, 7 and 8 
give the discharge of the “divisor”, “channel”, and the main ditch, 
respectively. To convert the discharge in second-feet, as given 
in columns 6, 7 and 8 into “inches” of water, multiply those values 
by the number of “inches” equivalent to a second-foot under the 
conditions in question. The quantity of water which an “inch” 
represents is sometimes different under neighboring canal sys¬ 
tems.* 
A dam placed across the divisor box at the point of the par¬ 
tition board increases the accuracy of the division of the water, 
its effect being most decided for low heads. When the velocity 
of flow was practically the same in both ditches, an effect on the 
division of a maximum of only approximately 1 or 2 per cent was 
caused by substituting a 6-inch dam for a 4-inch dam. A 
long box is more accurate than a short box, because the long box 
tends to equalize the velocities across the width of the flume be- 
*See Colorado Experiment Station Bulletin No. 207. 
