SORGHUM EXPERIMENTS ON THE GREAT PLAINS. 61 
Throughout these experiments it will be noted that the highest 
forage yields were obtained from the June 1 and June 15 seedings. 
The quality of the fodder was also considered superior to that pro- 
duced by seeding at an earlier date, because being harvested later it 
did not dry out so much before the winter feeding period. The data 
on seed production are less consistent and conclusive; but the best 
general results, excepting only feterita, were obtained by seeding 
about two weeks earlier than the optimum dates for forage. 
In preliminary tests with three varieties of kafir and with Dwarf 
Yellow milo during the period from 1907 to 1909, April 15 and May 1 
were the best dates for grain production, sometimes outyielding by 
more than 100 per cent the May 15 and June 1 dates. In 1907 the 
Blackhull kafir yields varied rather uniformly from 37 bushels on the 
April 15 date down to 5 bushels from the July 1 seeding. For Red 
kafir the yields varied in the same way, from 34 bushels to 10 bushels 
per acre. The milo yields in 1907 were not as uniform as those of the 
kafirs; the May 1 date ranked first, April 15 second, and June 15 
third. In 1908 the kafir yields were irregular, but showed best 
results from April 27 to May 15, with yields from 15 to 33 bushels. 
On June 1 kafir yields varied from 15 to 18 bushels, and all seedings 
made after June 1 failed to produce grain. Milo in 1908 yielded 22.5, 
27.9, 25.9, and 30.2 bushels per acre from the April 27, May 1, May 15, 
and June 1 seedings, respectively, and as in the case of kafir failed 
entirely on later dates. In 1909 there was no effective rainfall in 
July, August, and September, and the kafir varieties produced no 
grain yields on any of the dates. Milo in 1909 produced 14.8, 12.7, 
and 6.4 bushels per acre from April 15, May 1, and May 15 seedings. 
respectively, followed by failures on the later dates. 
A survey of the forage yields in 1907 and 1908 shows them highest 
on the June 1 and June 15 dates for both kafir and milo; the difference, 
however, between the early and late seedings was slight. In 1909 the 
early seedings gave markedly higher yields than the late ones, due to 
the low rainfall in the latter part of the season. The value of the 
forage was perhaps greater for the early dates in 1907 and 1908, owing 
to the larger quantity of grain it contained. 
The grain yields" obtained in the date-of-seeding experiments 
from 1913 to 1917 did not confirm the superiority of the early dates. 
Grain yields were highest for all the grain sorghums in seedings made 
from May 1 to June 1, owing in part at least to the intensive culture 
given the late-seeded plats from early spring until the time for seeding 
arrived. 
EXPERIMENTS AT AMARILLO, TEX. 
The experiments to determine the best date for seeding sorghums 
at Amarillo were conducted during the same years as at Chillicothe. 
Tex., but do not include the same varieties. Kafir and milo were 
omitted from the tests at Amarillo, in order not to duplicate work 
being done there with these varieties by the Office of Cereal Investi- 
gations (10). Results with Freed sorghum during the first three 
years were so disappointing on account of bird damage that Club- 
head, a promising early sorgo, was substituted for it in 1916 and 1917. 
As the entire season of 1913 was so adverse on account of drought 
and as the plats, although triplicated, were only 8-rod rows on which 
stands were made irregular by ground squirrels, the results offer but 
