310 ANNUAL REPORTS OF DEPARTMENT OF AGRICULTURE. 



in this rc<rion. Tlie results of these trials confirm tliosc of our own 

 experiments in showincf the yield of these ^Yheats to be much greater 

 tliun that of the varieties connuouly grown. 



Grain work in the southwestern plains. — drain work is done 

 on the experimental farms at Amarillo and Dalhart, Tex. The super- 

 intendent of the farm at Amarillo comes in contact with farmers to 

 a con>ideral)le extent at diU'ei'ent times during the year and is thus 

 able to disseminate among them, through institute lectures and other- 

 wise, a considerable amount of information obtained from the results 

 of our experiments. Over a large portion of this region the season 

 has been the driest of many years, thus giving an unusual oppor- 

 tunity to show the superiority of proper methods of cultivation over 

 those heretofore commonly practiced by the farmers. In all this re- 

 gion winter grains continue to show themselves superior to spring 

 grains, except in the case of oats. The short-season varieties of oats 

 are better than tliose tliat require a longer period for ripening, and 

 there is continued proof of the great importance of the sorghums as 

 dependable crops for profit. 



Influence of environment on the composition of grain. — The 

 environment experiments comprise several series, each consisting in 

 the planting of the same original seed of a variety of grain at three 

 different points and the transfer of enough of the succeeding crop 

 for seed from each point to each of the other two points the following 

 season for the purpose of determining the effect of a change of soil 

 and climate on the composition of the kernel. The past two 3'ears' 

 work in this line has given interesting results. In some of these 

 experiments there has been a departure from the results obtained 

 heretofore from wheat, in that the yield from home-grown seed has 

 not been as great as that from seed of the same original source 

 brought from other localities, though the dilTerence was not great. 

 In the soil-transfer experiments conducted in Mandand, Kansas, and 

 California the quality of the wheat produced has been practically the 

 same on each of the three different soils at the same point, though the 

 appearance of the growing grain in the three plats was quite different, 

 and the yields have usually been greater on the western soils. 



Production of pure strains of cereals. — In recent 3'ears a critical 

 examination of experimental work throughout the country has shown 

 that many experiments heretofore conducted with considerable pains 

 over a long jjeriod of time have been of little value because of the 

 neglect to observe certain fundamental facts in inaugurating the 

 experiments. One of these essentials is the employment of a pure 

 strain of the crop. The Office of Grain Investigations, recognizing 

 this fact, has given much labor to the sorting of so-called standard 

 and introduced varieties, which are really mixtures, and to the 

 development thereby of pure types from a single mother plant. This 

 work has now been carried on far enough for the acquirement of a 

 considerable amount of seed of a number of pure strains. As rapidly 

 as possible the seed of these strains will be substituted for the ordi- 

 nary seed heretofore used in all of the important field experiments. 



Crops in rotation avith cereals. — The crop-rotation experiments 

 intended to determine what crops should be used in alternation with 

 cereals to obtain the best results have been conducted for several years 



