8 BULLETIN 1067. U. S. DEPAKTIMENT OF AGRICULTUBE. 
The showing made by a pumping plant will depend to some extent 
on the amount of vegetation in the water and consequently on the 
time of year the test is made. A screen in the suction canal to keep 
weeds from reaching the pump is a necessity. Even where there is a 
good screen, weeds of small size will pass through and be caught on 
the blades of the impeller of the pump. The effect is to reduce both 
capacity- and efficiency. There is urgent need for some form of cutter 
that may be operated while the pump is in use. Such a device has 
been applied to one form of screw pump. It consists of a heavy cyl- 
inder of metal that is forced in and out by a hydraulic piston and so' 
placed that the blades of the impeller barely clear; any trash caught 
by the blades is thus sheared off and passes on through the pump. 
The patent involved also' covers the application of the device to 
centrifugal pumps, but so far as observation extends it has not been 
applied to that type of pump. 
Some pumping plants are operated at an improper number of 
revolutions per minute because of lack of data regarding the proper 
speeds for different lifts. Without a series of tests to determine 
the best speed of rotation a plant may be operated at considerable 
disadvantage. Because of the limited time ordinarily devoted to 
such a series, and the many limitations affecting the outcome of the 
tests, it is quite probable that the results do not represent the best 
performance of the class to which the plant belongs. 
The tests described hereafter were run by W. B. Gregory and J. M. 
Robert, of Tulane University, and C. W. Okey, Senior Drainage En- 
gineer, United States Department of Agriculture, assisted by B. S. 
Nelson, Charles Kirschner, and several members of the senior class 
in mechanical engineering at Tulane University. 
TE3T OF DRAINAGE WHEEL ON THE SOUTH SIDE PLANTING CO.'S TRACT, 
NEW ORLEANS, LA. 
This test was made in 1905 on a drainage wheel used to drain 
1,700 acres. The wheel was typical of its class, but had distinct 
features in the double gearing and in the number of paddles. Care 
was exercised so to design the wheel that the water would not be 
lifted unnecessarily. Its diameter was 28 feet and its width 6 feet. 
It was driven by a simple noncondensing engine of the slide-valve 
type, with a cylinder 16 inches in diameter and a stroke of 24 inches. 
The method of testing consisted in traversing the discharge flume 
with a current meter and taking indicator cards and other observa- 
tions as quickly as possible after the traverse was finished. By this 
means the indicated horsepower was a little less than the mean cor- 
responding to the water measurement, but as the latter required only 
about 10 minutes the error was not great. 
