4 CIRCULAR 558, U. S. DEPARTMENT OF AGRICULTURE 
DEAWNING GRASS SEED 
EFrect oF DIFFERENT MILL SPEEDS AND RATE OF FEEDING IN 
DEAWNING CANADA WILD-RYE SEED 
The preliminary deawning trials were made with seed of Canada 
wild-rye. This seed has long, stiff, curved awns, which make it 
bulky; consequently the test weight is low, 5 to 7 pounds per bushel. 
The tendency of this grass to ball when handled makes planting it with 
ordinary machinery almost impossible. A typical sample is shown in 
figure 1, A. In the lot chosen 93.5 percent of the seed was awned 
and hence was well suited for the deawning tests. The swinging- 
hammer mill, detailed specifications for which are given in table 1, 
was used. It was driven by a gasoline power unit, the speed of which 
could be regulated by means of a throttle. Seed was fed into the 
mill by hand. 
The initial trial, made at 2,250 revolutions per minute (fr. p. m.), 
the normal grinding speed of the mill, with a %-inch screen, gave 
eround seed. The screen was removed in the next trial. Without the 
screen, 50 percent of the seed was ground, and the remainder was not 
deawned. The use of a %-inch screen increased the amount of ground 
seed to 75 percent; the remaining 25 percent retained the awns. 
Provision was made to reduce the speed of the mill by installing a jack 
shaft between it and the motor. The pulleys used were capable of 
reducing the speed approximately one-half, so that with throttle 
control on the motor the mill could be operated accurately and 
smoothly at any desired speed between 480 and 1,120 r.p.m. It was 
found that a mill speed of about 700 r. p. m. with a ¥-inch screen gave 
a high percentage of deawned seed and a low percentage of injured 
seed. It was also observed that the rate of feeding affected the 
quality of the seed. 
After the preliminary trials, tests were begun to measure the effect 
of mill speed and rate of feeding on deawning. For each test a mini- 
mum of 30 pounds of seed and 3 minutes of operation were found 
necessary to obtain reliable data. The speeds of the hammer-muill 
cylinder were set and maintained with a Starret speed indicator. Two 
rates of feeding were used; the full rate, being that required to operate 
the mill at capacity without reducing the speed, and the half rate, at 
approximately half the capacity. Each trial was made in triplicate, 
care being taken to collect treated seed from the mill only while it was 
in full, continuous operation. This procedure assured representative 
samples and reduced error. The milled seed was cleaned, and aliquot 
samples were taken to measure the effect of the method of operation 
on the criteria shown in table 2. The data for three speeds of opera- 
tion and the two rates of feeding are given in this table. Only averages. 
of the three trials are given because of the slight variation obtained 
when one lot of seed was used throughout and the same operator fed 
the mill. 
A mill speed of 600 r. p. m. deawned 96 percent of the seeds and 
injured only 1.12 percent of the caryopses when the seed was fed at 
the capacity of the machine. Operating the mill at higher speeds 
increased the percentage of injured seed but had little effect on the 
percentage of seed deawned. Feeding the mill at approximately one- 
