SUFFICIENCY OF DATA SUBMITTED. 



13 



Bureau, which keeps a record of the physical and meteorological 

 conditions at each station. Several other offices of this Bureau are 

 also in cooperation at most of the stations. The Agriculturist in 

 Charge of Dry-Land Agriculture Investigations and his traveling field 

 assistant make frequent visits to the field stations during the growing 

 season for the purpose of inspecting the work. Other cooperating 

 heads of offices also visit these stations at frequent intervals. 



The work of seven of the eleven stations herein mentioned was con- 

 ducted in close cooperation with the state experiment stations of Mon- 

 tana, North and South Dakota, Nebraska, and Kansas, respectively, 

 and representatives of these stations are constantly in close touch with 

 the field work at their respective localities. Very full permanent 

 records of all the work are kept, both at the field stations and at the 

 central office at Washington, D. C. Both these records and the field 

 work itself are open to the inspection of the public at all times, and such 

 inspection is invited. The experiment stations mentioned use these 

 records in the preparation of publications of their respective stations. 

 With so many trained investigators actively interested in the elimina- 

 tion of all sources of error or inaccuracy, it seems reasonable to claim 

 that the data herein presented are trustworthy and that, covering as 

 they do so extensive an area, they constitute as reliable a basis for 

 safe conclusion concerning the methods of dry-land agriculture in the 

 Great Plains area as can now be found. It is only by a further 

 extension of these investigations that a more reliable basis can be 

 established. 



All of the eleven stations from which the data are collected are 

 located within what is generally known as the semiarid area. The 

 meteorological records and the crop yields under ordinary methods 

 indicate, however, that in thirteen seasons out of eighteen the weather 

 conditions were sufficiently favorable to give some basis for the 

 suspicion that these thirteen experiments were conducted under 

 humid rather than semiarid conditions and that they could not 

 therefore be used as a safe basis for conclusions relative to semiarid 

 conditions. When, however, we consider that the average increase in 

 yields from summer tillage as compared with ordinary methods during 

 these thirteen seasons is practically identical for oats and barley and 

 differs by only 2.2 bushels for wheat from those of the remaining five 

 seasons when the drought was severe, the force of this argument is 

 considerably weakened. If we also consider that the only instance 

 where summer tillage increased the yield sufficiently to make the 

 practice profitable was at North Platte in 1908, where conditions 

 were very favorable for crop production under ordinary methods, the 

 argument against the applicability of these results to semiarid con- 

 ditions ceases to have any weight." 



187 



a See Table XXXI for precipitation record. 



