CEItEAL INVESTIGATIONS AT THE NEPHt SUBSTATION. 39 



It is of interest to note that the thickest seedings resulted in the high- 

 est yields in all cases; also, that the common variety was generally 

 better than the durum variety, 



WINTER-WHEAT VARIETIES. 



The two varieties of winter wheat used for tests during 1910 to 

 1912, inclusive, were Koffoid (C. I. No. 2997) and Turkey (C. I. No. 

 2998). The Koffoid variety is beardless, with light-brown chaff and 

 large, soft, white kernels. The Turkey variety is bearded, with 

 white chaff and medium, hard, red kernels. Some difficulty was 

 experienced in grading these varieties, as was the case with the 

 spring wheats. A summary of the results obtained is presented in 

 Table XVI. The essential difference between this table and the 

 preceding one is that this is a summary of three years while the other 

 was a summary of but one. The average length of heads, also, has 

 been added in Table XVI. The figures for each year represent an 

 average of three rows. The averages, then, represent an average of 

 three rows for three years. In the right-hand column are given for 

 each size of seed the average results for all rates of seeding for three 

 years. A summary of the averages presented in the table is also 

 given, which will facilitate the making of comparisons. 



Table XVI shows that the percentage of survival from the different 

 sizes of seed varied widely during the 3-year test. The relative value 

 of the different grades would change each year if based on percentage 

 of winter survival. This behavior resulted in averages which showed 

 no great differences. The medium-sized seed of the Koffoid variety 

 gave the highest average, while with the Turkey the small seed gave 

 the highest average. The original unseparated seed of the Koffoid 

 was low in percentage of survival, while that of the Turkey was high. 

 The large seed of both varieties gave a relatively low percentage of 

 survival. In average number of plants maturing, the relative value 

 of the different grades was about the same as in percentage of 

 survival. 



In average number of heads per plant the large seed showed the 

 greatest value. In this respect the large seed of both varieties led 

 in almost every instance. The small seed and the original bulk seed 

 were practically equal in average number of heads produced per 

 plant, and both grades were lower in average than the medium size 

 of seed. The rate of seeding was an effective factor in determining 

 the number of heads per plant. This was lowest where the plants 

 were 2 inches apart in the row. Between the 2-inch and the 6-inch 

 rates of seeding there was a gradual increase in the number of heads 

 per plant. But after the 6-inch rate the number of heads per plant 

 remained about the same for all subsequent rates. From this it seems 

 that the 6-mch rate allows sufficient space in the row for winter wheat 

 to reach its greatest efficiency in tillering. 



