FIELD CROPS. 633 



and had been poorly farmed. For the period of the experiment winter wheat 

 was gi-own on the land continuously. The different methods of soil preparation 

 were as follows : Double disking without plowing, plowing 3 in. deep September 

 15, double disking July 15 and plowing 3 in. deep September 15, double disking 

 July 15 and plowing 3 in. deep August 15, single listing July 15, listing July 

 15 and splitting the ridges August 15, plowing 7 in. deep July 15, plowing 7 in. 

 deep August 15. plowing 7 in. deep August 15 and not working the land until 

 September 15, plowing 7 in. deep September 15, and plowing 3 in. deep July 15. 



The ground plowed each season on July 15 7 in. deep and thoroughly worked 

 after plowing produced as an average of three years a yield of 28.58 bu. of 

 grain per acre as compared with 24.94 bu. per acre on ground plowed August 

 15, and 13.62 bu. per acre on ground plowed September 15. Ground plowed 

 3 in. deep in July produced 10.92 bu. less wheat than ground plowed 7 in. deep 

 the same date. Both single and double listing gave smaller yields of wheat 

 than deep plowing. 



Early disking was advantageous when the plowing was delayed until Sep- 

 tember, but no increa.se in yield resulted when the plowing was done in August. 

 Regardless of method, early working usually produced an increase and late 

 working a decrease in yield. The soil moisture content at seeding time seemed 

 to bear practically no relation to the yield of wheat. The amount of soil 

 moisture at seeding time was found to be increased more by the size of the 

 preceding crop than by the method of preparing the gi'ound. 



Early plowing and thorough working of the soil were found to liberate large 

 quantities of nitrates, and in 1911 and 1913 there was a close correlation 

 between the amount of nitrates in the soil at seeding time and the yield of 

 wheat. While not in direct proportion, yet large yields were only obtained when 

 nitrates were present in large quantities and small yields were usually secured 

 when nitrates were low. Moisture was more completely used by the crops on 

 the plats containing a large supply of nitrates at seeding time, and there was 

 considerable evidence to show that moisture was more economically used in the 

 presence of a large supply of nitrates. Nitrogenous fertilizers did not increase 

 the yield of wheat on ground rich in nitrates plowed in July, while on disked 

 corn ground the nitrogenous fertilizer increased the yield of wheat nearly 100 

 per cent. 



The general conclusion is drawn that apparently early plowing in preparing 

 land for winter wheat is of value principally because of the large supply of 

 plant food liberated, especially nitrates, rather than from any other cause, 

 and that poor results are secured from late plowing chiefly because plant food 

 is not made available in sufficient quantities. 



The progressive development of the wheat kernel, II, R. W. Thatcher 

 (Jour. Amer. Soc. Agron., 7 (1915), No. 6, pp. 273-282). — This article reports 

 the continuation of work previously noted (E. S. R., 31, p. 234). 



The varieties of wheat used in these trials were of the Fife, Velvet Chaff, 

 and Blue Stem groups. Harvesting was begim with each group as it reached 

 a particular stage of development, 1. e., when it showed the largest proportion 

 of spikes with the anthers of the two central spikelets protruding. Samples 

 were taken every three days. The data show the yield and physical properties 

 of kernels at the successive stages of development ; analyses of sample kernels ; 

 weights in milligrams of material per kernel ; gain in milligrams of protein and 

 carbohydrates in each kernel, with the ratio of the two materials for each 

 period ; the relation between the composition of material gained by the kernels 

 and their specific gravity ; and the character of the nitrogen-containing com- 

 pounds in Blue Stem wheat at successive stages of growth. 



