FACTORS IN CROP PRODUCTION 581 



ranging from 11.21 to 17.09. It is not surprising, perhaps, that 

 " dry farming " should succeed fairly well in recent years with a 

 rainfall ranging from 1 8 to 28 inches, and there is reason enough 

 to convince many 1 that rainfall follows the plow. However, the 

 heaviest rainfall on record is for 1883 (29.88 inches), and the aver- 

 age for the ten years, 1877 to 1886, is only .38 of an inch below the 

 average for the last ten years, while the average for the first 17 

 years is .64 of an inch greater than the average for the last 

 17 years, according to the records of the 34 years. 



A matter worthy of important consideration is the distribution 

 of rainfall. A rainfall of 15 to 20 inches is sufficient for very fair 

 crops if it comes at the rate of 3 inches a month from April to 

 September; but, if two or three torrential showers of 4 or 5 inches 

 each all within a month or six weeks are parts of the total, the rain- 

 fall may be very inadequate. 



The author is firmly of the opinion that most of the cultivable 

 semiarid lands in the United States where the average annual 

 rainfall exceeds 15 inches should be and will be occupied, and also 

 that a very satisfactory measure of prosperity can be attained by 

 those who farm those lands under the best methods; but it should 

 not be forgotten that there are certain to be periods of severe drouth 

 sometimes for several years in succession; and, unless adequate 

 provision is made against such times, there will be suffering for 



1 An experience reported to the author by Mr. N. S. Spencer, a resident of 

 Champaign County, Illinois, cannot fail to be of interest, and may be of some value, 

 to students of semiarid agriculture. Mr. Spencer stated that he went into central 

 Nebraska some years ago and saw growing in the fields wheat crops that yielded 

 35 bushels per acre on very low-priced land, and he had positive assurance that ex- 

 cellent crops had been grown the year before and also in previous years. He bought 

 a large farm, and broke up 400 acres the same season, on which wheat was 

 seeded in the fall. The following year crop failure was common, and he threshed 

 no wheat. However, there were some good summer rains and he prepared the 

 land well and again seeded 400 acres of wheat, but again the rain failed and 

 he threshed no wheat. Once more the summer rains were sufficient to enable 

 him to put the land in good condition, and he sowed 300 acres of wheat, which, 

 however, also resulted in complete failure. He then gave up the land upon which 

 he had made two payments, disposed of his stock and tools as well as he could, 

 and found that his total loss for the three years' experience amounted to about 

 $10,000. 



Soon after hearing this story, the author looked up the rainfall record as reported 

 above for North Platte, and then stated to Mr. Spencer that he must have bought 

 his land in 1892, which was found to be correct. 



