8 BULLETIN 1011, U. S. DEPARTMENT OF AGRICULTURE. 
In 1918 the experiment was repeated at the field station at Rock- 
ville, Md. The results, presented in Table I, show that the field 
stands two weeks after planting varied from 75 per cent of a perfect 
stand for the dehulled seed to 98 per cent for the check. No data were 
obtained on the time of pollen shedding, but judged from the moisture 
content of the ears at harvest the mutilation of the seed had retarded 
the maturity of the plant. The height of plants at maturity could 
not be correlated with the treatment of the seed. . The average weight 
of ears in the mutilated classes was less than in the adjacent checks 
with one exception. The number of ears to the plant and the average 
yield for each plant were less than in the adjacent checks in most of 
the comparisons. 
The experimental plantings at the Yarrow Plant Introduction 
Field Station in 1919 were destroyed by floods shortly after planting. 
ARLINGTON EXPERIMENT. 
The data obtained in the experiment at the Arlington Experimental 
Farm in 1919, presented in Table I, show that good stands were obtain- 
ed. No counts were made, however, preceding the thinning. The 
dates of first pollen shedding were later in the mutilated-seed classes 
than in the check. The time of ripening did not appear to be mate- 
rially affected. The general height of plants of the different lots at 
maturity did not noticeably differ. The weight of ears in the muti- 
lated seed classes was less than from the adjacent checks except in 
one case. In number of ears to the plant, the dehulled class exceeded 
the adjacent checks in both comparisons, the starchless tied the check 
in one instance and was less in the other, and the germs were less than 
the checks in both cases. In average yield to the plant the muti- 
lated classes were less than the adjacent checks with one exception. 
EXPERIMENTS ON THE EFFECTS OF THE MUTILATION OF THE GERMS 
In the experiments conducted in 1920 at the Arlington Experi- 
mental Farm the germs as well as the endosperms were mutilated. 
The -seeds were cut into longitudinal and transverse sections. The 
longitudinal sections were made by cutting through from the germ 
side to the back of the kernel, dividing it approximately into halves 
(PI. Ill, fig. 2) ; the cross sections were cut transversely to the long 
axis of the embryo, dividing the kernel into crown and tip portions. 
The crown portion contained much of the endosperm, part of the 
scutellum, and a small portion of the plumule (PL III, fig. 4); the 
tip contained a smaller amount of the endosperm, part of the scutellum, 
all of the radicle, and a portion of the plumule (PL III, fig. 3). 
Preliminary investigations conducted in the laboratory were con- 
fined to germination tests and to observations during a short period 
of the early growth of the plants. In the field experiments, the 
