EXPERIMENT STATION BULLETINS. 



625 



loss of water under arid and humid conditions, Buckiugliam endeavored 

 to imitate these two conditions in the laboratory. He placed soil in 

 cylinders 48 inches long and 21^ inches in diameter, and provided each 

 cylinder with side tubes at the bottom for the introduction of water. He 

 allowed, by means of an electrical fan, a current of air to be blown over 

 the top surface of the soils. For the arid conditions this current of air 

 was heated, without changing its absolute humidity, to a temperature of 

 about 50° to 60° F. above the room temperature. To imitate also the 

 high surface temperature of soils under the strong sunshine of arid cli- 

 mates, the top inch and a half of the cylinders under the hot air was 

 heated, by heating coils surrounding the cylinders, to about the same 

 temperature as the hot air. The breeze of about 3 miles per hour, was 

 kept going all the time. The heating current was turned on for six hours 

 a day except on Sunday and holidays. For the humid conditions the 

 soils were placed under the current of air at room temperature. Bucking- 

 ham performed a number of experiments bearing upon this subject and the 

 results he obtained are qualitatively about the same for all of them. The 

 following diagram shows a typical set of results. 



>300 



Z>/))<5 et-f^iBD 



FIG. 5. CURVE SHOWING EVAPORATION OF WATER FROM PODUNK FINE SANDY 

 LOAM WITH TAP WATER; A, SOIL UNDER HUMID CONDITIONS; B, SOIL UNDER 

 ARID CONDITIONS: C, WATER UNDER ARID CONDITIONS; D, WATER UNDER 

 HUMID CONDITIONS. (AFTER BUCKINGHAM.) 



An examination of the above figure shows that the loss of water from 

 the soil under arid conditions is much more rapid at first but after 

 about 4 days have elapsed the rate of loss is less under arid than under 

 humid conditions, and continue to be so throughout the duration of the 

 experiment. The rate of evaporation from the soils for the last ten days 

 is 11.2 inches of rain per year under arid conditions and 51.6 inches of 

 rain per year under humid conditions. 



Buckingham explains these results under the supposition that a mulch 

 was formed on the soil kept under arid conditions, more rapidly than on 

 the soil kept under humid conditions, and the mulch prevented rapid loss 

 of water from the former. This explanation is correct, of course, in so 

 far as it represents the result of the mulch, but how this mulch was 

 formed and how it was capable of accomplishing this result, it fails to 



