52 DULLETIN 1115, U. S. DEPARTMENT OP AGRICULTURE. 
varietal experiments were sown on one date each year, and the plan 
has been to make this date the same from year to year. It so hap- 
pened that the date on which the sorghums in the varietal experi- 
ments were seeded proved to be the best for Sunrise kafir, but a rela- 
tively poor date for Dwarf Yellow milo. According to the date-of- 
seeding results with the two varieties, Dwarf Yellow milo when seeded 
on Jun3 15 has outyielded Sunrise kafir when sown on the date most 
favorable for that crop by an average of 6.4 bushels per acre annually. 
SPACING EXPERIMENTS. 
Spacing experiments have been conducted with Dwarf Yellow 
milo (C. I. No. 332) and Sunrise kafir (C. I. No. 472) during the 5-year 
period from 1917 to 1921, inclusive. The objects of these experiments 
were to determine the best distance between plants and between 
rows and to study the effect of the different spacings on these grain 
sorghums. These experiments consist of two divisions for each of the 
varieties. The first division included five plats with rows spaced 44 
inches apart, the plants being spaced at different distances in the row 
in each of the five plats. The second division consisted of five plats 
with rows spaced 88 inches apart. The distance between plants in the 
row in the second division was half as great as in the corresponding 
plat of the first division, thus giving the same number of plants per 
unit area in the corresponding plats in the two divisions. 
In these spacing experiments the milo and kafir plats were seeded 
on the same date in the first three years, these dates being May 22 
in 1917, May 27 in 1918, and May 21 in 1919. During the last two 
years the milo and kafir in the spacing experiments were seeded on 
different dates. In 1920 the kafir pla'ts were seeded on May 17 and 
the milo plats on May 26, while in 1921 the kafir pkits were seeded on 
May 17 and the milo plats on June 10. In conducting spacing ex- 
periments the plats should be seeded on the date which gives maxi- 
mum yields for the variety. This was possible only during the year 
1921, as the date-of-seeding results were hardly conclusive until 1920. 
However, the Sunrise kafir in the spacing experiments has been 
seeded during the period which the date-of-seeding experiments 
show is most favorable for the production of grain. The Dwarf 
Yellow milo in the spacing experiments was seeded too early to give 
maximum grain yields in 1917, 1918, 1919, and 1920, but seeding on 
June 10 gave very high yields in 1921. 
To obtain uniform and comparable stands in the spacing experi- 
ments it has been necessary to drill the seed very thick and then 
thin by hand to the desired distance between plants. This method, 
though impracticable for farming operations, has proved satisfactory 
for this experimental work. Thinning should be done as soon as 
the plants are large enough (4 to 6 inches high) to pull without 
breaking at the first node. The spacings desired for the plats with 
rows 44 inches apart were one plant for each 6, 12, IS, 24, and 30 
inches of row space and for the plats with rows 88 inches apart one 
plant f©r each 3, 6, 9, 12, and 15 inches. Though these spacings 
were not obtained each year, they have been approximated, as the 
data show. 
For convenience the spacings have been designated as shown in 
Table 22. 
