DRY FARMING IN SOUTHEASTERN WYOMING 17 
were May 1, 30.6 bushels; May 15, 29 bushels; and June 1, 26 bushels. 
With both varieties there was a sharp decrease in yield from sowing 
as late as June 1, but little choice as to the time of seeding before the 
middle of May. 
Kherson oats were sown at the rates of 3, 4, 5, 6, and 7 pecks per 
acre in each of the five years from 1914 to 1918, inclusive. The 
5-year average yields from each of these rates were 33.3, 36.6, 
39.3, 40.6, and 38.4 bushels per acre, respectively. These results 
indicate the best rate of seeding to be from 5 to 6 pecks per acre. 
A 2-rowed barley, Svanhals (C. I. No. 187), was used in an experi- 
ment on the date of seeding barley which was conducted for the 
seven years from 1913 to 1919, inclusive. The dates of seeding 
were April 15, May 1, and Ma} 7 15, except in 1918, when there was 
no seeding on April 15. The 7-year average 3d elds for the May 
1 and May 15 dates of seeding were 18.8 and 20.4 bushels per acre. 
Excluding 1918 the 6-year average acre 3d elds from each of the 
three dates of seeding were April 15, 15.1 bushels; May 1, 15.4 
bushels; and May 15, 15.7 bushels. It is evident that the best 
time to sow barley is during the first half of Ma3 T . The ill effect 
of late seeding was shown in 1920 when Coast barley sown on 
disked corn ground in the varietal tests on May 22 yielded 20.8 
bushels per acre and sown on the same preparation in the rotation 
experiments on June 14 3d elded only 5.9 bushels. 
Rate-of -seeding tests were conducted with Svanhals barley for 
the seven years from 1913 to 1919, inclusive. The average yields 
from seeding at the rates of 2, 3, 4, and 5 pecks were 19.9, 22.2, 23.7, 
and 22.9 bushels per acre, respectively. It is evident that about 4 
pecks per acre is as much seed as is necessary. 
The results of seeding flax at different dates depends more upon the 
cleanness of the seed bed from weeds than it does upon the date 
itself, provided the date is early enough to allow the crop to mature. 
In the date-of-seeding experiments the practice was to double-disk 
the land to be seeded to flax the day before seeding. This gave the 
late seedings the cleaner seed beds, as the weeds in them had germi- 
nated and were destroyed before seeding. The experiments were 
conducted both on disked corn ground and on fallow. The better 
yields were on disked corn ground. The average acre yields for 
seedings on April 15, May 1, May 15, and June 1 were 4.8, 
6.6, 7.5, and 8.2 bushels, respectively. In 1918 a seeding made on 
June 15 3d elded 8.4 bushels per acre. A seeding as late as this 
usually would not mature. 
The o-jear average yields of flax sown at the rates of 10, 15, 20, 
and 25 pounds per acre were 8.9, 9.6, 9.7, and 9.5 bushels per acre, 
respectively. The 10-pound rate appears to be too light, but the 
differences between the other rates did not material^ affect the 
3delds. The variety used in both the date-of-seeding and the rate-of- 
seeding experiments was Russian (C. I. No. 19). 
The problems in the dates of seeding and rates of seeding of winter 
wheat are peculiar to winter cereals. The chief difficulty in the 
production of winter wheat is soil blowing. The effect of the date 
of seeding on this factor is very important. If a good growth is 
produced before cold weather, the crop is certain to pass the winter 
in good condition unless it is damaged by soil blowing, and the better 
