443 



were coufiiieil to ii clay soil underlaid with sand. Some of tbe bearings of 

 these phenomena on the tillage of this class of s<mIs are briefly discussed. 



(1) Cultivation after rains. — Uuless tho ground is already too wet the stirriug of the 

 surface soil, wherever practicable, shouhl follow just as soon after a considerable 

 rain-fall as the tools will work well. Tho cultivation should, as a ruh^, bo shallow, 

 leaving a thin stratum of the surface soil finely pulverized and completely cut off 

 from the ground below. If this is not done, the extrenndy rapid evaporation which 

 takes place from undisturbed wet soil on hot, clear days may, even in a i\i\v hours, 

 not only dissipate that wlii(;h has just fallen but also a part of that which the rain 

 has caused to be drawn toward the surface from lower levels, and thus leave the 

 ground actually drier, aa a whole, than before the rain, even though it may look 

 more moist at the surface. 



(2) Watering ti-ans pi anted trees. — When dry weather follows the planting of trees 

 it will be evident that simply wetting the surface may, in certain localities, do more 

 harm than good, because, in these cases the roots, lying as they do at considerable 

 depths, can not use water which remains at the surface, and as surface wetting may 

 diminish the water content of the deeper soil, the soil about the roots is liable to be 

 rendered drier than before the wetting. * » * 



If, however, the surface soil about the trees is deeply spaded before watering, the 

 water will then enter the ground more deeply by the direct force of gravitation, 

 largely unimpeded by capillary action, while at the same time the ability of the soil 

 to return the water to the surface will be reduced to the minimum, and if a good 

 mulch is now added the water will be under the best conditions for being used by the 

 tree. So, too, if the soil about the roots of transplanted trees is well firmed to insure 

 the rapid transit of water to them, while the surface is left loose and well mulched 

 at the time of setting to prevent capillary action upward above the roots and to 

 permit the rains to penetrate downward to them, we start the tree under the best 

 possible conditions for growth, so far as moisture is concerned. 



(3) Relation to irrigation. — The bearing of these facts upon problems of irrigation 

 will be evident to those directly interested, and need not be dwelt upon here. 



Capillary movement of icater in field soil in its natural condition and 

 position (pp. 139-145). — (1) Cases of slow movement upward. — In the 

 Annual Eejjort of the station for 1889, p. 202 (See p. 434 above), are 

 recorded observations indicating a veryslowcapillarymovementof water 

 in the soil of cultivated fields. Observations in the same line in 1890 

 are reported in this article. From sami)les of soil taken from the same 

 field as in the previous experiments, it was determined " that the total 

 gains of water in the upper 4 feet of soil iu the field during 85 days 

 [December 13, 1889. to March 8, 1890] were only 1.04 per cent of the 

 weight of the dry soil, or an aggregate of 3.07 pounds. Rain-fall during 

 the interval was 5.18 inches, or 26.98 pounds per square foot." In the 

 same way it was shown " that the capillary movement upward into the 

 upper 4 feet of soil under consideration during the 41 days from March 

 8 to April 18, 1890, when coupled with a rainfall of 21.77 pounds of 

 water per square foot, was only sufficient to increase the water of these 

 4 feet 11.38 pounds, barely one half of the rain-fall itself." Combining 

 the observations during about G months from October, 1889, to April, 

 1890, it appeared that the lowest of the 4 feet of soil under consideration 

 had increased its water content only 3.32 per cent, or 3.1 pounds per 

 cubic foot, though at no time during this period wg,s it more than 3^ 



