14 BULLETIN 99, U. S. DEPARTMENT OF AGRICULTURE. 



Cluster, was used. This is a standard variety for the State of New 

 York. These check rows, however, were not used in computing the 

 yields reported herein, which are the actual 3?ields of the various 

 strains. 



SOIL CONDITIONS. 



The tests have been made on a soil of the Dunkirk clay-loam 

 type, except in 1908, when the planting was made on a gravelly loam. 

 In 1910 a duplicate series of tests was also grown on the gravelly 

 loam in order to see whether there was any difference in the results 

 from the two types of soil. As the experiments have progressed, 

 an effort has been made to make the conditions uniform. Ordinary 



Fig. 5. — Side view of the machine used in thrashing rows of oats at the Cornell University experiment 



station. 



cultural conditions were duplicated as nearly as possible; hence, only 

 moderate applications of fertilizers were made. 



YIELDS OF GRAIN FROM THE SELECTIONS. 



The selections which produced the highest yields per acre in the 

 6-year test were 34al-ll-2 from the hybrid Sixty-Day X Burt, with 

 an average yield of 61 bushels; selection 123-5 from the commercial 

 variety Welcome, 60.5 bushels; selections 50al-10 and 50al-22 from 

 the hybrid Sixty-Day X Probsteier, 60.4 and 58.2 bushels, respectively; 

 and selection 62— II— 1 8—1—1 from the commercial variety Sixty-Day, 

 57 bushels. Of the selections which exceeded 55 bushels in yield in 

 the 6-year test, 4 were from series 34, Sixty-Day X Burt; 4 from 

 series 49, Sixty-Day X Clydesdale; 2 from series 50, Sixty-Day X Prob- 



