500 STATE BOARD OF AGRICULTURE. 



uniform moisture content in brass tubes 8 inches long and liA inches 

 in diameter, closing both ends with solid rubber stoi)pers, and keeping 

 one-half of the soil column at a high temperature and the other half at a 

 low temperature for a certain length of time, then determining the per- 

 centage of moisture of the two columns, and aUrihuling any difference in 

 water content to tliermal translocation. There were only two amplitudes 

 of temperature employed, 0° to 20° C, and 0° to 40° C, i. e., one-half of 

 the soil column was kept at 0° C. and the other half at 20° and 40° C. 

 For producing these temperature amplitudes wooden boxes were used 

 which contained melting ice and Avarni water separately and their tem- 

 peratures were maintained constant by the addition of ice and hot water 

 respectively. 



The movement of moisture from warm to cold soil Avas studied in two 

 different ways: (1) when the column of soil stood horizontally, and (2) 

 when it stood vertically. For the first case, the wooden boxes used were 

 22 inches long, 10 inches wide, and 20 inches deep, having a wooden parti- 

 tion in the center Avhich contained perforations of the size to fit the tubes. 

 One compartment contained melting ice and the other water at the de- 

 sired temperature. To prevent any exchange of Avater between the two 

 compartments the edges of the partition and the holes through which the 

 tubes passed were made water-tight by means of paraffin. For the .second 

 study, the employment of two boxes was necessary. One box, which con- 

 tained melting ice, was 24 inches long, 10 inches wide, and 13 inches 

 deep. The other box, which contained water at the desired temperature, 

 was made to fit the inner side of the first box and was 13 inches long, 7 

 inches wide and 11 inches deep. The bottom of the small box was sup- 

 plied with holes in which to place the tubes. The size of the perforations 

 was made to fit the tubes tightly. To make the small box also water 

 proof the crevices surrounding the tubes were sealed with melted paraffin. 

 The small box was then put upon supports in the large box. The latter 

 was filled with melting ice up to and touching the bottom of the former. 

 The water in the upper box was kept at the desired temperature. A 

 diagram of both set of boxes is shown in figure 1. All the boxes were 

 well insulated, and since they were big and contained large volumes of 

 water the temperature could "be kept to within small variations for long 

 time. The water was stirred occasionally, to maintain uniformity of 

 temperature throughout the mass. 



The temperature amplitudes employed are within the upper limit of the 

 diurnal amplitudes of temperature at the upper depths in the soil but 

 they are too high for the range of temperature that exists at any one 

 time between the various adjacent depths. 



The duration of each experiment was about eight hours. This time 

 limit was calculated to represent approximately the length of time 

 that the day and night soil temperature gradient is most marked. 



The effect of temperature on the movement of moisture in soils of uni- 

 form moisture content was investigated in 5 diverse classes of soil, 

 namely: Miami light sandy loam, Miami heavy sandy loam, Miami silt 

 loam, Clyde silt loam and Miami clay. Each soil contained a large num- 

 ber of different moisture contents. These various moisture contents in 

 each soil ranged from very low to very high. 



To procure very uniform moisture content throughout the soil column 

 each soil, after it was moistened to the desired degree, was passed through 



