The precariousness of dry-farming must be done away with. The year of drouth must be expected every year. Only as certainty of crop yield is assured will dry-farming rise to a respected place by the side of other branches of agriculture. To attain such certainty and respect clean summer fallowing every second, third, or fourth year, according to the average rainfall, is probably indispensable; and future investigations, long enough continued, will doubtless confirm this prediction. Undoubtedly, a rotation of crops, including hoed crops, will find an important place in dry-farming, but probably not to the complete exclusion of the clean summer fallow. Jethro Tull, two hundred years ago, discovered that thorough tillage of the soil gave crops that in some cases could not be produced by the addition of manure, and he came to the erroneous conclusion that "tillage is manure." In recent days we have learned the value of tillage in conserving moisture and in enabling plants to reach maturity with the least amount of water, and we may be tempted to believe that "tillage is moisture." This, like Tull's statement, is a fallacy and must be avoided. Tillage can take the place of moisture only to a limited degree. Water is the essential consideration in dry-farming, else there would be no dry-farming. CHAPTER XI SOWING AND HARVESTING The careful application of the principles of soil treatment discussed in the preceding chapters will leave the soil in good condition for sowing, either in the fall or spring. Nevertheless, though proper dry-farming insures a first-class seed-bed, the problem of sowing is one of the most difficult in the successful production of crops without irrigation. This is chiefly due to the difficulty of choosing, under somewhat rainless conditions, a time for sowing that will insure rapid and complete germination and the establishmcnt of a root system capable of producing good plants. In some respects fewer definite, reliable principles can be laid down concerning sowing than any other principle of important application in the practice of dry-farming. The experience of the last fifteen years has taught that the occasional failures to which even good dry-farmers have been subjected have been caused almost wholly by uncontrollable unfavorable conditions prevailing at the time of sowing. Conditions of germination