WATER LOSSES FROM WET AREAS 21 



experiments were carried on in triplicate to average errors and the 

 effects of soil differences. The following: snmraaries and conclusions 

 are given from the data obtained during the investigation : 



1. ^Mariotte tanks Avere nsed with all soil tanks to supply and main- 

 tain a constant water table in the soil. Their value lies in the ease 

 with which periodic measurements of water used may be made as they 

 are automatic in operation. Great care is necessary to protect the 

 ^Mariotte tank against temperature changes or from leakage of air 

 into the tank or the connecting pipe system. 



2. Evaporation tests from bare soil were conducted with both dis- 

 turbed and undisturbed soil in tanks, separately. No evaporation 

 occurred from tanks of undisturbed soil having a water table 4 feet 

 below the surface. When the water table was raised to 3 feet from tlie 

 soil surface the evaporation averaged 0.1 acre-inch per acre per month. 

 Tanks of undisturbed soil having water tables at a depth of 2 feet lost 

 an average of 0.445 acre-inch per acre per month for 15 months. In 

 contrast with losses from undisturbed soil, three tanks filled with dis- 

 turbed soil having a 2-foot depth to water level had a mean monthly 

 loss by evaporation of 1.599 acre-inches per acre, while the average 

 loss from undisturbed soil for the same period was 0.404 acre-inch 

 per acre, or about 25 per cent. In disturbed soil the opportunity for 

 evaporation was greater as the soil contained more moisture. Evapora- 

 tion from undisturbed soil is more comparable to that lost under field 

 conditions than is that from disturbed soil. These data indicate that 

 there will be no evaporation from the light textured soils of the Hanford 

 series when the water table is 4 feet or more below the ground surface. 



3. Use of water by both salt grass and Bermuda grass was influ- 

 enced by the availability of moisture in the soil and the depth to water 

 table. Grasses in tanks having the highest water tables used the most 

 water. During the year ending April 30, 1932, salt grass grown in 

 tanks having water tables 1 foot in depth used water at the rate of 

 42.75 acre-inches per acre; with a 2-foot depth, 36.23 acre-inches; and 

 with a 5-foot depth, 22.12 acre-inches per acre. In general, the ratio 

 of the use of water to depth of water table b}* Bermuda grass was about 

 the same as that of salt grass. From these data, it is concluded that 

 use of water by these grasses is not excessive and does not exceed the 

 amount that would be used by many cultivated crops grown under the 

 same conditions of soil moisture. 



4. Consumptive use of water bj- tules or cat-tails grown in tanks 

 in exposed locations is not closely indicative of the true use by these 

 plants growing in their natural environment. Growths in exposed 

 tanks are subject to greater solar radiation, lower humidity and greater 

 wind movement conditions than are found under natural swamp condi- 

 tions; and use of water by swamp growth transplanted to exposed 

 locations is inordinately high. Numerous instances of tules in tanks 

 using an acre-inch or more of water per 24 hours at the Santa Ana 

 station and an extreme use of 3.6 acre-inches per acre per 24 hours 

 at Prado were noted. Taken as a percentage of evaporation from a 

 standard Weather Bureau pan the use of water by tules or cat-tails in 

 exposed tanks varied from 168.3 per cent for cat-tails at Santa Ana to 

 451.7 per cent for triangular stem tules at Prado. From other experi- 

 ments bv the Bureau of Agricultural Engineering it is evident that 



