34 BULLETIN 30,U. S. DEPARTMENT OF AGRICULTURE. 
The length of these heads and the number of kernels each contained 
were determined. Finally 30 kernels from each head, except when 
the head contained fewer kernels, were planted in rows 10 feet long 
and 12 inches apart. The kernels were sown 4 inches apart in the row 
at the usual time, both in spring and fall. A furrow about 3 inches 
in depth was first opened with a wheel hoe. Then a board set with 
wooden pegs 4 inches apart was used to mark the places for the seeds. 
The seeds were placed by hand, one in each hole left by the pegs. 
The furrows were then covered with the wheel hoe and a rake. This 
left the seed from 3 to 4 inches beneath the surface of the ground. 
From then until the resulting plants were mature their behavior was 
observed and noted. Data pertaining to the following points were 
collected: Date of full emergence from the ground, percentage of 
survival, color, vigor, manner of growth, culms per plant, heads per 
plant, and height. Then what was believed to be the best plant 
growing under normal conditions in each row was selected for further 
propagation. The points considered in making these selections were: 
Vigor, uniformity of culms, number of heads, and length of heads. 
After the plant had been selected the remaining plants in the row 
were harvested and thrashed. The weight of their grain with that 
of the selected plant gave the total row yield. The mother plant 
selected was treated in the same manner as the parent plant of the 
previous year. This procedure has been repeated each year since 
1908. During that time several selections of more recently introduced 
varieties have been added to the original number. The nursery at 
present contains selections of 95 varieties of winter cereals, including 
wheat, oats, barley, emmer, and spelt; and 23 varieties of spring 
cereals, including only oats and durum wheat. In only a few instances 
are there fewer than two selections of any one variety and in many 
instances there are more than two selections, sometimes as many 
as 20. ; 
From such a large number of head rows, recorded as they are at 
Nephi, a great mass of data has been accumulated. Not all of these 
are germane to the purposes of this bulletin. Those which are of 
interest, however, in connection with the results obtained in the 
plat tests of the same varieties will be discussed here. 
WINTER CEREAL HEAD ROWS. 
An unexpectedly high percentage of winterkilling has occurred in 
the winter cereal head rows. The average winter survival has been 
about 65 per cent. This has varied, of course, with the severity of 
the winter conditions, and in that respect the rows were affected in 
practically the same proportion as were the plats. 
The tillering of the plants has varied with the thickness of the 
stand. Where the stand was thick the plants have grown more erect 
