258 SOILS. 



of absorbing any moisture from a wet brick, while a dry 

 brick will readily withdraw nearly all the water contained in 

 the relatively large pores of the sponge (see chap. n). A 

 layer of loose, dry surface-soil is therefore an excellent pre- 

 ventive of evaporation of the moisture from soils, and may 

 be regarded as the natural and most available means to be 

 used by the farmer, both for the prevention of evaporation 

 and to moderate the access of excessive heat and dryness to 

 the active roots. 



As regards the desirable thickness of this protective layer 

 of tilled surface-soil, it should be kept in mind that in the 

 humid region, where rain can be expected at intervals of from 

 one to three weeks, the feeding roots may usually be found 

 within a few inches of the surface; while in the arid region, 

 where irrigation is practiced at long intervals or sometimes 

 not at all, so that no water enters the soil oftener than from 

 two to six months, the roots necessarily vegetate at lower 

 depths, and hence the protective surface-layer can, and should 

 be, of greater thickness, to prevent the penetration of excessive 

 heat and dryness during the long interval. 



The failure to appreciate this necessary difference often 

 leads to heavy losses on the part of newcomers to the arid 

 region, who in this as in other respects are apt to follow 

 blindly the precepts familiar to them in the East, until taught 

 better by sore experience. In the East and Middle West a 

 depth of three inches is considered the proper one for the pro- 

 tective surface-layer; and in the case of maize even this is 

 considered excessive in many cases. In the arid region this 

 depth should be at least doubled where irrigation is not prac- 

 ticed at least every four to six weeks ; and in some sandy soils 

 even seven and eight inches is not too much for effective 

 protection. 



Illustrations of Effects of Surface Tillage. The efficacy of 

 loose surface tilth in preventing evaporation, as compared 

 with mere superficial scratching or with the total omission of 

 cultivation, is well exemplified in a series of investigations 

 conducted on this subject during the extremely dry season of 

 1898, by the California Experiment Station; the seasonal rain- 

 fall having during that year been on an average from one- 

 third to one-half onlv of the usual amount, so as to test to the 



