12 



BULLETIN 49S, U. S. DEPARTMENT OF AGRICULTURE. 



every fifth plat sown uniformly to one variety and regarded as a 

 check. The two methods described in previous publications 1 have 

 boon used to determine relative or computed yields, but neither lias 

 proved entirely satisfactory, The variation in check-plat yields of 

 spring grains usually has not been great.* Occasionally, however, 

 wide variations occur in the yields of check plats which can not be 

 explained in any satisfactory manner. 



Fig. 6.— A bar wecder, or "slicker," in operation at the Eastern Oregon Dry-Farming Substation. 



Beginning with 1913, each variety of spring grain in a varietal test 

 has been sown in duplicate twentieth-acre plats. The plats in the 

 two series are arranged as follows: 



1 



11 



2 

 12 



3 

 13 



4 

 14 



5 

 15 



(i 

 16 



7 

 17 



8 

 18 



9 

 19 



10 

 20 



11 

 1 



12 

 2 



13 

 3 



11 

 4 



15 



5 



16 

 6 



17 

 7 



18 



8 



19 

 9 



20 

 10 



All yields in this bulletin are reported in bushels per acre, based on 

 the actual yields of a single tenth-acre or the average actual yields of 

 two twent Let b-acre plats. 



TREATMENT OF PLATS. 



The general practice in growing cereals in the Columbia Basin is to 

 alternate a grain crop with bare fallow, commonly called summer 



! Cardon, P. V. Cereal investigations at the Nephi substation. U. S. Dept. Agr. Bui. 30, p. 12, 33. 

 1913. Clark, J. Allen. Cereal experiments at Dickinson, N. Dak. U. S. Dept. Agr. Bui. 33, p. 11, 12. 

 1914. 



