26 BULLETIN 336, U. S. DEPARTMENT OF AGRICULTURE. 



When the difference in the rate of seeding is considered, the net 

 yields of the Dietz from the 3-peck and 8-peck rates were practically 

 the same, 24.20 and 24.10 bushels, respectively, with the other rates 

 of seeding averaging from 1 to 2 bushels lower. The net yields of the 

 Fultz show that the 4-peck rate gave the best average result, 30.85 

 bushels, with a consistent reduction when less or more seed was sown, 

 except that the 8-peck rate gave a higher net yield than any of the 

 others except the 4-peck and 5-peck rates. 



*?he Stoner gave the highest net yields from the 3-peck and 4-peck 

 rates, 23.75 and 23.95 bushels, respectively. The net yields from the 

 other rates of seeding were from 2 to 3 bushels lower. The highest 

 net yield from Martin Amber, 22.6 bushels, was obtained from the 

 3-peck rate, with decidedly lower net yields from all the other rates 

 except the 4-peck. 



The average for all four varieties which were grown at the seven 

 rates of seeding in 1913 and 1914 show slightly higher net yields from 

 the 3-peck and 4-peck rates than from any of the others. The net 

 average yields from these rates are 24.41 and 24.61 bushels, respec- 

 tively. There is a constant reduction in the average net yield from 

 higher and lower rates of seeding, except that the 8-peck rate gives 

 a yield of 0.9 bushel higher than the 7-peck rate. 



Although these tests have not been conducted long enough for the 

 data to be conclusive, it appears that quite as good or perhaps slightly 

 better yields may be obtained when wheat is sown at the rate of 3 or 

 4 pecks on fertile, well-prepared land than when sown at rates ranging 

 from 5 to 8 pecks to the acre. As there is a decided saving of seed 

 at the lower rates, these are to be preferred where conditions are 

 similar to those existing on the Arlington Farm. It is not known 

 how widely these results are applicable. 



EXPERIMENTS WITH WINTER SPELT AND EMMER. 



Spelt and emmer are closely related to wheat, but the spikelets do 

 not break up in thrashing, as in wheat, the kernel remaining inclosed 

 in the glumes. When the glumes are removed, the kernel is quite 

 similar to that of wheat, not only in appearance but in composition. 

 These grains are used chiefly as feed for domestic animals. Because 

 the inclosing glumes give considerable bulk to the grain, the feeding 

 value of spelt and emmer is about the same as that of oats or barley. 

 These, grains, particularly spelt, are not grown as generally in the 

 South Atlantic States as experimental results seem to warrant. 



Varietal tests of spelt and emmer have been conducted at College 

 Park since 1908 and at Arlington since 1910. The tests at College 

 Park have included five varieties of spelt and two varieties and one 

 selection of emmer. Only two varieties of spelt, one of emmer, and 

 the emmer selection, however, have been grown for the entire six 





