THE FLOW OF WATER IN IRRIGATION CHANNELS. 7 
of course, remains the same throughout the reach tested, but as a rule 
the mean area, A, is not identical with the area at the place where the 
discharge was measured, so V will not necessarily be the same as the 
mean velocity at the section where the discharge was measured. 
The other field data to be taken in order to make the resulting 
value of n fully comprehensible is a careful description of the mate- 
rial forming the containing channel, including such growths as affect 
the flow of the water and a description of the influence of all struc- 
tures hi the canal and all changes in alignment throughout the reach 
tested. This general description should not only cover just the reach 
tested, but should extend up and down stream for sufficient distances 
to cover anything influencing the flow within the reach. 
Temperatures of the air and water may be taken, but it is doubtful 
if any deductions may be made as to the direct influence of the vari- 
ous temperatures on the flow of water in the usual more or less 
irregular channel. 
SCOPE OF EXPERIMENTS. 
Tests were made on channels in Nebraska, Colorado, Utah, Idaho, 
Oregon, Montana, California, Arizona, Texas, and Louisiana. These 
channels ranged in size from small ditches carrying less than 1 second- 
foot up to canals carrying over 2,600 second-feet. The containing 
materials of these channels comprise wood, concrete, earth, rubble 
masonry, cobblestones, and a few special combinations. The veloci- 
ties encountered extend up to about 10 feet per second. From other 
sources the writer has obtained the data for additional tests, where in 
his opinion there was not sufficient evidence in our own experiments 
from which to draw conclusions. This is particularly true of steel 
flumes, as none of these visited by members of this force was carry- 
ing water at the time. The data covering very high velocities, such 
as are found hi chutes, also came from outside sources. In several 
cases it was possible to get data covering several tests on exactly the 
same reach of channel, with varying discharges of water, with a view 
to proving or disproving the theories of some writers who have con- 
tended that the value of n diminishes as the discharge increases, in 
proportion to some function of the latter, such as the velocity, the 
hydraulic radius, or the square root thereof. 
EQUIPMENT AND METHODS EMPLOYED FOR COLLECTING FIELD DATA. 
In order to weigh correctly any new data advanced which tend 
either to corroborate or to change existing elements in a standard for- 
mula it is necessary to know in detail the instruments employed and 
the methods of taking field measurements. Consequently, the equip- 
ment used and the steps pursued in developing the values of n for 
irrigation canals for various conditions are discussed in some detail 
in the following paper. The equipment and the methods described 
