July i6, 1917 Movement and Distribution 0} Moisture in Soil 



137 



Figure 18 represents the relative weekly losses in the wet soils and 

 gains in the drier soils of this experiment. During the first week the 

 gain increased with the increased initial percentage, but thereafter the 

 reverse was true, very little gain taking place in the soil having 8.24 per 

 cent of water after the first week. 



DETERMINED BY THE DEFLECTION OP ONE END OP A SMALL COLUMN OP SOIL 



This experiment is essentially a duplication of the last one except that 

 6-inch test tubes were used instead of 8-inch colorimeter tubes and that 



ReLeiii^'e disfribalicn of moisfure of yar/ou^ times 

 in i^et section in driy section 









I 



R 



L/ne, of separation beiween wei and drq soiLi 



Fig. 19.— Diagram showing the effect of varying initial percentages of water on the horizontal distribu- 

 tion of moisture in drier soils in contact with a wet soil having 28.44 per cent of water. The top line of 

 each series represents the original distribution. 



the standard wet soil contained 28.44 per cent of moisture, while the 

 drier soils had 7.40, 8.68, 10.30, 12.54, 13.08, and 14.79 per cent of water, 

 respectively. This test lasted from April 5 to May 22, 191 6, the record 

 being made as in the previous experiment. 



In figure 19 are given the relative weekly losses in the wet and gains 

 in the drier soils. Although the soils with initial percentages of 8.68 

 and 12.54 are exceptions, the gain in weight for the first week decreased 

 with an increase in the initial percentage. After the first week the gain 

 decreased more uniformly with the increasing initial percentage. 

 98974°— 17 3 



