EXPERIMENT STATION BULLETINS. 533 



EFFECT OF TEMPERATURE ON THE RETENTION OF WATER BY 



SOILS. 



Since the amount of water retained by soils is believed to depend to 

 a considerable extent upon the surface tension of the soil water, and in- 

 asmuch as we have seen that the surface tension of water is reduced by 

 increase in temperature, it was decided to conduct an investigation to 

 obtain numerical values of these relations. This research was facilitated 

 by the fact that the same soils of the last section could be used. ' After 

 the studies on percolation were completed these soils were brought to 

 the temperature of melting ice and allowed to remain at this temperature 

 for about 24 hours to drain. Their temperature then was raised to 50° 

 C. by increments of 10° C, and any water that dripped down during a 

 definite period at any particular temperature was caught and weighed. 

 Preliminary trials showed that during a period of two hours the soils 

 allowed to drain practically all the water that would drain at any one 

 of the temperatures employed. In fact, at the higher temperatures, the 

 soils ceased draining even before the completion of two hours. Through- 

 out this investigation, therefore, the limit of two hours was used for all 

 temperatures and for all soils. Table 6 and the accompanying figure 9 

 show the results obtained. The table contains also the percentage of water 

 drained and thereby lost, at the different temperatures, based upon the 

 dry weight of the soil. 



The results show that in the case of the sandy loam, silt loam, clay 

 loam, clay and muck the amount of water drained increases regularly and 

 rather rapidly with increase in temperature. In the case of the sand, 

 however, the quantit}^ drained increases up to the temperature of 30° C. 

 and then commences to diminish with further rise in temperature. The 

 total quantity drained is about the same in the former soils, and about 

 three times greater than in the latter soil. 



The order of these results complies, in a general way, with what might 

 have been expected, but whether the magnitude of the values, which is 

 rather appreciable, is correct, cannot positively be asserted, because the 

 relation of soil to water and other factors, complicate the problem. Ac- 

 cording to the figures presented in table 1, the surface tension of water 

 decreases at the rate of about 0.2 per cent per degree Centigrade, so that 

 for every 5° C. the rate decreases about 1 per cent. From this it would 

 appear that the decrease in the water holding capacity of the soils fol- 

 lows the law, but not closely. It is certain, however, that the above 

 quantities of water drained are not wholly due to the decrease in the sur- 

 face tension of the soil water, but also to two other factors. These are 

 (1) the expansion of the air in the interstices, and (2) the expansion 

 of the soil mass, and consequently to the squeezing out of water. As the 

 water leaves the soil by drainage at the temperature of melting ice, its 

 space is replaced by air. When the temperature of the soil is raised this 

 air, some of which is imprisoned, expands, and produces a sudden pres- 

 sure throughout the soil mass and forces out or expels some of the free 

 water which accumulates at the lower end of the tube, due to the greater 

 attraction of gravity. As a proof of this may be cited the fact that the 



