110 I 



Each tank was equipped with automatic water-supply Mariotte 

 apparatus. In some respects these were unsatisfactory, as they 

 were not protected against temperature changes. In periods of 

 rising temperatures, expanding air forced water out of the bottle 

 beyond the capacity of plant absorption, and the water table in 

 the soil tank rose above the desired level. As air in the bottle 

 became cooler, flow of water was retarded and transpiration oc- 

 curred faster than water could be supplied. Water levels dropped 

 in the soil tank. Over a considerable period of time these 

 changes were unimportant, but they destroyed the opportunity to 

 obtain accurate hourly records of transpiration losses. 



Greasewood . --Attempts were made to separate transpiration 

 from greasewood plants and evaporation from soil. Separate soil 

 evaporation tanks were used for this purpose. Saltgrass growing 

 in tanks shades the ground surface so that little soil evaporation 

 occurs, and the principal loss is caused by transpiration. In a 

 tank of greasev/ood there would be some bare soil, and evaporation 

 would be a factor in the total loss of water. White (38) has 

 estimated this as approximately 25 per cent of the total. Con- 

 sumptive use of water by saltgrass and greasewood grown in tanks 

 in the Escalante Valley is presented in Table 31. 



SAN LUIS VALLEY, COLORADO 



Saltgrass . --Tipton and Hart, for the State Engineer of 

 Colorado, conducted studies for several years on the use of 

 water by saltgrass in tanks and of evaporation in the San Luis 

 Valley, Colo. (32) . 



The evaporation and transpiration laboratory was established 

 at Garnett in 1927 and continued in 1928. The station was rehabil- 

 itated and placed in operation again in April 1930 and continued in 

 1931. No change was made in the apparatus or in the depth at which 

 the water table was maintained in the various tanks. An additional 

 saltgrass tank was installed to maintain the water table at a depth 

 of about 40 inches, but this tank did not begin to function properly 



