WATER LOSSES FROM WET AREAS 53 



althoug'li sufficient data were collected to obtain monthly totals through- 

 out the following spring. Some weekly totals were omitted. Because 

 of changes in depths to water tables in October, 1930, true records 

 could not be obtained pending soil moisture adjustments and data for 

 these periods are not included. At other times accident or failure of 

 the Mariotte tanks to function properly is responsible for omissions. 



A complete summary of all work done at the Santa Ana station 

 for each of the three years of the investigation is given in Tables 13, 14 

 and 15, which show the monthly use of water by all soil and water 

 tanks. Descriptions of contents of tanks are included, and percentages 

 of use of water by each moist area growtli with reference to evaporation 

 from water surfaces is likewise tabulated. These data are the most 

 valuable of the report. 



Use of Water by Bermuda Grass 



Bermuda grass {Cynodon dactylon) is a perennial with long, creep- 

 ing, jointed stems, often several feet in length. It spreads largely by 

 rooting at the nodes, although it also seeds abundantly. Where condi- 

 tions are favorable it forms a dense turf, frequently becoming a pest in 

 lawns by driving out or smothering other laAvn grass. It is found groov- 

 ing wild in many localities, always in exposed places, as it is intolerant 

 of shade. Bermuda grass is not an indicator of ground water, as is 

 salt grass, although the experiments indicate that it may use slightly 

 more water than salt grass when it is available. It is frequently used 

 for pasture and makes good feed for stock. 



Excellent conditions for experimenting with Bermuda grass were 

 found at the San Bernardino station. The yard in which the tanks 

 were set was covered with a dense growth of the grass, so that tests of 

 consumptive use of water were made with tank crops surrounded by a 

 natural growth of the same variety. The principal difference between 

 tank and field conditions was in depth to water table. In the tanks, the 

 water table was fixed at definite levels during the three years of record, 

 while the outside ground water fluctuated between 2^ and 6 feet below 

 the ground surface. 



Data on weekly use of water by Bermuda grass grown in tanks 

 at San Bernardino are given in Tables 16 and 17. In November, 1931, 

 daily readings at the station were discontinued, because of high ground 

 water entering the tanks through the waste pipes and changing the 

 water levels. Records of weekly and monthly use of water immediately 

 before and after the first week in October, 1930, are not comparable 

 as at this time the grass was maliciously burned off from all four tanks. 

 Previous to the burning, each tank had a heavy crop of grass which 

 used over \ acre-inch of water per week. Immediately after the burn- 

 ing, this loss was reduced over one-half. As the burning was done in 

 the fall, no recovery was possible until the following spring when new 

 growth a])i)eared on all tanks. In spite of the loss of grass, there was 

 a small but continuous loss of water from each tank throughout the 

 winter months. Plate VI shows the appearance of the grass in the 

 tanks and in the surrounding field before burning took place. Tank 

 growth is showni in the center of the picture as being higher than the 

 surrounding grass. 



