EXPERIMENT STATION BULLETINS. 517 



temperature at night with rise in elevation is called surface temperature 

 inversion. At this minimum temperature the atmospheric pressure ap- 

 proaches its maximum, and the inward flow of air is thereby facilitated. 



All the foregoing facts lead to the enunciation of a general law that 

 during the day, as the tempcratwe rises, the soil air tends to floio out- 

 ward into the atmosphere, and during the night, as the temperature 

 falls, air from the atmosphere tends to floio inivard into the soil. This 

 law diametrically opposes the prevalent theory that during the night 

 there is an upward movement of moist warm air. The above law, how- 

 ever, seems to be borne out by logic and appears to be confirmed by ex- 

 perimental evidence subsequently to be presented. The prevalent theory, 

 however, seems unreasonable. For instance, if it is admitted, which it 

 must be, that the soil air escapes into the atmosphere during the day as 

 the temperature rises, then where and when does the soil obtain its air 

 if it continues to give up air even during the night? It might be argued 

 that it is vapor that is rising to the surface and not air. That is incon- 

 ceivable in the present case. It is true that distillation would occur if 

 the amplitude of temperature were appreciable and constant, but it has 

 been shown that the temperature of the whole column of soil decreases 

 constantly, and that an air current from the cold atmosphere is drawn 

 inward which tends to encounter and oppose any upward movement of 

 vapor rising from any difference in temperature. Moreover, granting 

 for sake of argument that there is a vapor rising from the warmer soil 

 to the colder soil at the surface, the amount would be extremely small 

 to account for the great quantity of dew commonly noted, because the 

 temperature amplitudes of the soil at different depths at night are never 

 very great. In fact, during the spring months as the temperature of 

 the lower depths continually rises and the trend of the air temperature 

 is upward, the range of temperature between the surface and the lower 

 depths say 4 inches, is small, usually amounting only to about 2 or 3° C. 

 The greatest differences in temperature at the different depths in the 

 morning occur in the fall when the trend of the air temperature is down- 

 ward and the surface soil temperature continually falls. At this time 

 the variation in temperature between the surface and six inches of the 

 mineral soils may be as high as 8° C. but the variation between adjacent 

 depths is only from about 0.5° to 2° C. 



The truth of the matter, however, seems to he that histead of vapor 

 rising from the icarmer soil deloio to the colder soil at the surface, vapor 

 enters the soil from the atmosphere. This is a natural conclusion from 

 the law enunciated that during the dav air is exhaled from the soil and 

 during the night air is inhaled from the atmosphere. The amount of 

 moisture that will thus enter the soil will depend upon the quantity of 

 air inhaled and upon its absolute humidity, but as will be seen subse- 

 quently it is extremely small. The water may be abstracted by the dry 

 soil at the surface as the air is drawn in or it may enter unaffected. 

 Thus, it is possible that the moisture lost by the soil during the day 6?/ 

 the expulsion of its moist air is partly, if not wholly, regained at night. 



What is then the source of water of the dew? The greatest part of 

 it comes from the lower layer of the atmosphere itself by condensation. 

 Some of it comes from the leaves of trees and plants. And a certain 

 amount comes from the soil by capillary and thermal capillary action 

 as set forth previously. 



According to the foregoing consideration, therefore, the notion that 



