510 STATE BOARD OF AGRICULTURE. 



the point of intersectiou, theu this theoretical curve agrees almost per- 

 fectly with the real one in figure 2. 



The serious fault with the above illustration is that the total effective 

 ])ull tends to become zero, and, theoretically, this should not be the case, 

 because while the pull due to (he attractive power of the soil for water, 

 and to the curvature of the capillary films will ultimately become zero, 

 the pull due to the increased surface tension of the soil water should 

 not become zero, but should remain the same for all moisture contents. 

 Hence, the above diagram illustrates more correctly only the thermal 

 translocation of the water as due to all the other forces except to the 

 surface tension of water. 



The next important question to consider is the mode and amount of 

 thermal translocation of water in field soils as suggested by the fore- 

 going laboratory experimental data. Under field conditions the soil 

 moisture exists practically always in a gradient form. As the water 

 content tends to decrease upwards from the water level, the forces clue 

 to the curvature of the capillary films, and to the attractive power of 

 the soil for water, increase correspondingly, consequently the pull is up- 

 ward. The soil temperature also exists in a gradient form, but this 

 reverses itself diurnally and therefore modifies these pulling forces. Dur- 

 ing the day the temperature at the upper depths is higher than that be- 

 low, the attractive and adhesive forces of the soil for water, and the sur- 

 face tension of water, are decreased, so that the total upward effective 

 pull is diminished correspondingly. Inasmuch as the temperature be- 

 low is less than that above, the effective pull due only to the increased 

 attractive and adhesive forces of the soil for water and to the surface 

 tension of the soil water, should occasion a dov/nward movement of mois- 

 ture. Since, however, the water attractive forces of the soil below are 

 more satisfied than those of the soil above, the dowuAvard pull, due only 

 to the attraction, adhesion and surface tension as increased by a lower 

 temperature, is very small in comparison with the upAvard pull. Hence, 

 during the day, the moisture movement is upward. During the night, 

 nearly all of the above forces act in a parallel direction and favor an up- 

 ward movement. Hence, the thermal movement of capillary moisture in 

 soils is ahvays upward and never dowmcard. 



The extent to which moisture will move during the night from the 

 warmer soil below to the colder soil above will depend, (1) upon the- 

 soil temperature gradient, that is, upon the difference in temperature of 

 the various adjacent depths, and (2) upon the gradient or amount of 

 moisture content at the various depths. In the preceding series of ex- 

 periments the temperature amplitudes of 20° and 40° C. were employed. 

 In nature, however, so large and sharp variations in temperature be- 

 tween adjacent depths never occur during the night; (they do occur, 

 however, at the uj)per depths between day and night). Soil temperature 

 investigations which iirc being conducted at this Station show that in 

 the early morning when the temperature gradient is most marked, the 

 temperature of the bare mineral soils increases, sometimes in the sununer 

 and fall at the average rate of about 2° or 3° C. for each inch depth 

 down to about 4 inches and then this rate becomes less. In cropped 

 soils, where the temperature remains more constant, this rate of in- 

 crease of temperature with depth is still less. Hence, the amount of 

 thermal translocation of wntcr that would occur during a single night 

 ivould he very small. On the other hand, the maximum thermal trans- 



