52 BULLETIN 462, IT. S. DEPARTMENT OF AGRICULTURE. 
The first experiments by the writer in Florida were made in a 
grove near Orlando, in 1909. The owner of the grove had set up a 
25-horsepower engine and a pump which did not work satisfactorily. 
He was advised to install a single-stage centrifugal pump, capable 
of delivering about TOO gallons per minute to the highest part of the 
grove, about 35 feet above the surface of the lake from which the 
water supply was obtained. Two 5-inch wrought-steel mains carried 
the water to the top of the grove, a distance of about 500 feet. 
The question now was how to distribute this water to the best 
advantage from the end of the main without expending large sums 
of money on iron or steel piping systems. Upon making a survey 
of the -iO-acre grove it was found that all of it could be covered by a 
pipe system requiring heads of water not larger than 10 to 15 feet 
after the water had been pumped through iron pipe to a high point 
in the grove. Concrete pipe has been found satisfactory in some 
parts of Florida, but the poor grade of sand which had to be used 
discouraged its general use there. The sand was procured from Lake 
Weir and is considered good for ordinary concrete work, such as 
building blocks ; but a sharp, clean sand should be selected for good 
concrete pipe, such as -is used for irrigation purposes in the West. 
The principal objection to this pipe was the difficulty, with the labor 
available, of concreting the separate joints together so as to make 
them watertight. This could be accomplished in several ways, but 
would require special apparatus and considerable care. 
Terra-cotta sewer pipe was found to answer every requirement, 
although care was necessary in laying it. (PL V, fig. 1.) The 
remainder of the distributing system was, therefore, finished with 
the terra-cotta pipe, which has given the best of satisfaction during 
the last six years, and judging from its use throughout both East and 
West it should continue satisfactory for an indefinite period. 
The outlets from the terra-cotta distributing system were made by 
connecting special irrigation valves to short upright pieces of 6-inch 
terra-cotta pipe, which were connected to the main pipe by the ordi- 
nary terra-cotta T. Six, eight, and ten-inch pipes were used. 
These were laid in trenches on the highest ridges throughout the 
grove. The pipe was cemented together so as to make water-tight 
joints, and was then covered with earth, the top of the pipe then be- 
ing at least 12 inches beneath the surface. 
Probably the most important item in the construction in this grove 
was the arrangement to prevent excessive pressure in the terra-cotta 
pipe. If the iron main from the pump were connected directly to 
the terra-cotta pipe it is probable that the pipe would burst from 
water hammer, or sudden starting or stopping of the water. To pre- 
vent any possibility of this a relief stanclpipe was constructed at the 
